Overview and Recommendations

Ninety-two thousand people were killed in the Tambora, Indonesia, eruption of 1815; over 820,000 were killed in the Sian, China, earthquake of 1556; 240,000 were killed by the earthquake of 1976 at Tangshan, China; and about 6000 were killed in the 1985 earthquake in Mexico City. Less spectacular, but economically serious, are the effects of slower tectonic processes or tectonic events of smaller amplitude that can, for example, disrupt river channels or warp coastlines and harbors. Only the most stable blocks of the Earth’s crust provide suitable long-term repositories for the disposal of radioactive waste, yet almost no region is completely free of tectonic changes over millenia. Thus, the evaluation of active tectonic processes is critical to many of mankind’s activities, so that hazards can be minimized and structures can be sited and constructed in ways that serve their functions most effectively, economically, and safely.

To fully evaluate ongoing tectonic activity and its associated hazards requires knowledge of the rates, styles, and patterns of tectonic processes. Many of these processes cannot be described reasonably using the limited instrumental or historical records; however, most can be described adequately for practical purposes using the geologic record of the past 500,000 yr.

In the following discussion active tectonics is defined as tectonic movements that are expected to occur within a future time span of concern to society, and the significance of active tectonism to society is of special concern. The entire range of geology, geophysics, and geodesy is, to some extent, pertinent to this topic; but the needs for useful forecasts of tectonic activity, so that actions may be taken to mitigate hazards, call for special attention to ongoing tectonic activity, its rates, styles, and patterns. The question of what constitutes ongoing is itself addressed because the patterns of activity leading to some catastrophic events cannot necessarily be delineated by a few-year-long data base or even by all of recorded history. For example, some of the largest earthquakes that have occurred were generated along segments of great faults that are now seismically very quiet. In some places, the historical seismic record suggests, inaccurately, an inverse pattern of what the long-term geologic record demonstrates.



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 3
Active Tectonics: Studies in Geophysics Overview and Recommendations Ninety-two thousand people were killed in the Tambora, Indonesia, eruption of 1815; over 820,000 were killed in the Sian, China, earthquake of 1556; 240,000 were killed by the earthquake of 1976 at Tangshan, China; and about 6000 were killed in the 1985 earthquake in Mexico City. Less spectacular, but economically serious, are the effects of slower tectonic processes or tectonic events of smaller amplitude that can, for example, disrupt river channels or warp coastlines and harbors. Only the most stable blocks of the Earth’s crust provide suitable long-term repositories for the disposal of radioactive waste, yet almost no region is completely free of tectonic changes over millenia. Thus, the evaluation of active tectonic processes is critical to many of mankind’s activities, so that hazards can be minimized and structures can be sited and constructed in ways that serve their functions most effectively, economically, and safely. To fully evaluate ongoing tectonic activity and its associated hazards requires knowledge of the rates, styles, and patterns of tectonic processes. Many of these processes cannot be described reasonably using the limited instrumental or historical records; however, most can be described adequately for practical purposes using the geologic record of the past 500,000 yr. In the following discussion active tectonics is defined as tectonic movements that are expected to occur within a future time span of concern to society, and the significance of active tectonism to society is of special concern. The entire range of geology, geophysics, and geodesy is, to some extent, pertinent to this topic; but the needs for useful forecasts of tectonic activity, so that actions may be taken to mitigate hazards, call for special attention to ongoing tectonic activity, its rates, styles, and patterns. The question of what constitutes ongoing is itself addressed because the patterns of activity leading to some catastrophic events cannot necessarily be delineated by a few-year-long data base or even by all of recorded history. For example, some of the largest earthquakes that have occurred were generated along segments of great faults that are now seismically very quiet. In some places, the historical seismic record suggests, inaccurately, an inverse pattern of what the long-term geologic record demonstrates.

OCR for page 3
Active Tectonics: Studies in Geophysics Understanding the period from the present to about 500,000 yr ago (the late Quaternary) forms the best basis for analyzing active tectonics of concern to society. To study this period of time calls for a special mix of geologic, seismological, geophysical, and geodetic techniques, many of which need major research efforts to achieve their full potential. Some of the most critical needs are highlighted in this Overview. Detailed discussions of recent advances in understanding active tectonic processes and their rates appear in the authored chapters that follow this Overview. Many research programs that can provide background for the study of active tectonics have been considered in earlier reviews by the National Research Council, including such reports as Geophysical Predictions (1978), Earthquake Research for the Safer Siting of Critical Facilities (1980), Geodetic Monitoring of Tectonic Deformation—Toward a Strategy (1981), Effective Use of Earthquake Data (1983), Seismographic Networks: Problems and Outlook for the 1980s (1983), Explosive Volcanism: Inception, Evolution, and Hazards (1984), and Seismological Studies of the Continental Lithosphere (1984). Research needs in, for example, seismological and geodetic techniques of importance to active tectonics are well formulated in these documents. The present study makes no attempt to review all these previous analyses but builds on them and specifically endorses some previous recommendations. For the most part, the previous reports have not addressed the premodern part of the time frame of the past few hundred thousand years, which must be understood to evaluate ongoing tectonic activity fully. Some potentially powerful techniques that should be developed by accelerated research are considered in this study. The recently emerging subdiscipline of paleoseismology—in which geologic techniques are used to identify and evaluate prehistoric earthquakes—has provided some of the most important recent advances in earthquake prediction. Similar techniques have also permitted evaluation of seismic hazards for urban areas and for critical facilities such as dams and nuclear reactors. Essential to such geologic research and evaluations are the ages of geologic units. Only by having the geologic history calibrated by known dates can we calibrate the (1) recurrence intervals of earthquakes and volcanic eruptions and (2) continuing rates and changes of rates of all tectonic processes. The determination of rates of processes and means of dating materials of late Quaternary age (past 500,000 yr) thus are considered of high priority for research attention. Among other facets of geology that have been underused in studying active tectonics is geomorphology. Landforms are everywhere; they are extremely sensitive to active tectonics; and geomorphic analysis has the potential for providing insights into active tectonic rates, styles, and patterns of deformation available through almost no other approach. Serious efforts are needed to accelerate research in quantitative geomorphology. The need is recognized for improving geodetic measurements of active tectonics through new land-based instruments, such as two-color laser distance-measuring devices, and space-related techniques, such as the Global Positioning System (GPS). Regional seismic networks are considered essential to track the patterns of ongoing strain release, to assist in mapping the activity of faults, and to assist in tracking the movement of magmas under volcanoes. Rapid data gathering, processing, and analyses are imperative. Great volumes of data from regional seismic networks and a variety of strain meters and geodetic networks must be handled quickly if useful forecasts (with lead times up to a few weeks) of earthquakes, volcanic eruptions, and landslides are to be made successfully. In summary, this study addresses tectonic processes, their rates, and methods of identifying and evaluating active tectonics by analysis of events, especially in the time frame from the present to about 500,000 yr ago. Except for brief comments, the socioeconomic and engineering accommodations for coping with the problems created by

OCR for page 3
Active Tectonics: Studies in Geophysics active tectonic processes are not addressed; they are in the domain of other specialists and other reports. ACTIVE TECTONICS In the earth sciences “tectonics” refers to the deformational structures and architecture of the outer parts of the Earth and to the evolution of these features through time. Examples include folds, warpings and tiltings of crustal blocks, and displacement on faults. In this study we define “active tectonics” as “tectonic movements that are expected to occur within a future time span of concern to society.” This definition of active tectonics has evolved from the many definitions of active fault that have been used in the past. The lack of agreement on a single definition of an active fault has caused confusion and attendant engineering, social, and legal difficulties. Many of the definitions inappropriately have mixed elements including criteria for identifying faults, criteria for estimating degree of activity, and value judgments about the level of activity that constitutes acceptable risk to mankind. The terms active faults and active tectonics imply that events are currently happening, but here the term currently is ambiguous. The present is a moving instant that progresses ever forward in time. To most people currently very likely is thought of in minutes, or at most in years, whereas to a geologist a time sample that appropriately represents currently or present might span many thousands of years. Focusing the definition of active tectonics on the future avoids selecting a single period of time to represent the present and emphasizes the prediction of future tectonic events, which has the greatest potential benefit in guiding hazard-reduction actions. The frequency of occurrence of specific faulting events has been incorporated in some definitions of active faults. Thus, various public agencies define an active fault as having had displacements (a) in 10,000 yr, (b) in 35,000 yr, (c) in 150,000 yr, or (d) twice in 500,000 yr. Such definitions have come to carry legal significance, and the great range in time frame has caused confusion. The differences reflected in these definitions relate principally to the degree of activity and to the levels of risk that are acceptable to various agencies. A far less confusing approach is to simplify the definition itself, to describe the rates of processes separately from the processes themselves, and to judge what risk is acceptable to society separately from the description of processes and their rates. Geologists, geophysicists, and geodesists can identify the geologic structures and can describe and evaluate the degree of activity and the patterns of activity. It is the role of engineers to try to accommodate and minimize the deleterious effects of tectonic activity; policymakers must decide whether the rates of processes and engineering accommodations of those processes result in an acceptable situation or level of risk. TECTONIC PROCESSES Tectonic deformation may occur as broad warping of the Earth’s surface, termed epeirogeny, to produce or reshape the larger features of continents and ocean basins, or it may be orogenic, that is, in more localized regions and belts to form mountain chains. The vertical movements of epeirogeny may result in plateaus and basins, whereas the more complex deformational processes of orogeny, or mountain building, include, for example, folding, faulting, plastic deformation, plutonism, and volcanism. The styles and rates of tectonic processes range widely; the most active and complex tend to occur along the margins of lithospheric plates. The continental margin of western North America, where the Pacific, North American, and Juan de Fuca-Gorda plates join, exemplifies the complexity that can exist at such boundaries. Strike slip predominates where the Pacific and North American plates are in contact in Califor-

OCR for page 3
Active Tectonics: Studies in Geophysics nia; convergence and subduction occur where the oceanic Juan de Fuca-Gorda plates slide beneath the continental North American plate in Oregon and Washington; and, where the three join at the Mendocino triple junction, several directions and rates of plate motion must be accommodated in a complex geometric pattern. The oblique relative movements of the Pacific and North American plates farther south in Mexico result in the shifting of Baja California northwestward away from continental Mexico. A similar pattern of crustal extension is present across the intraplate Basin and Range province. Examples of epeirogenic movements include the broad uplift of the Colorado Plateau, the Monroe Uplift across the lower Mississippi River Valley, the isostatic rebound of the Canadian and Scandinavian shields following deglaciation, and the depression of parts of the East Coast of North America. Active orogenic movements are well characterized by active faulting and active folding. Major strike-slip faults, such as the San Andreas Fault system in California, disrupt the landscape by offsetting streams, sedimentary basins, and mountain masses. The lateral movements are almost always accompanied by at least some vertical movement of crustal blocks. Irregularities in the fault trend, or on the fault surface, cause local rotation of stress; crustal blocks may be forced upward or downward—generally with the vertical movement accommodated along either reverse or normal faults—or the blocks may be rotated between branches of a fault. Extension of the Earth’s crust dominates in some provinces, and, as in the Basin and Range province of the western United States, extension is accompanied by vertical adjustments of range-sized blocks under the influence of gravity. Horsts and grabens develop as individual crustal blocks rise or drop, or the range blocks tilt so that one edge drops downward while the other edge rises. Many ranges along plate boundaries are the products of folding of stratified sediments. Folding generally represents processes of crustal shortening; crustal shortening also may involve the thrusting of one block over or under another, along relatively gently inclined faults. Where crustal plates collide rather than slide laterally past one another, the oceanic plate tends to slide beneath the continental plate in a process known as subduction, although not uncommonly part of an oceanic plate rises and slides over the continental plate (obduction). During collision with the Asiatic crustal plate, the Indian subcontinental plate has been thrust perhaps more than a thousand kilometers beneath the Himalayas and the Tibetan Plateau. Many smaller crustal blocks, or microplates, have collided with and have been accreted to the major continental plates. Investigations during the past decade suggest that some microplates have migrated at rates of tens of centimeters per year and have traveled across several tens of degrees of latitude. Such mobility of the Earth’s crust is an important aspect of active tectonics. In the overall crustal shortening process, large thrust faults emerge from the lower crust and flatten as they near the surface so that slabs of crust kilometers thick move tens of kilometers over underlying parts of the crust. In addition, large slabs (kilometers thick) of the Earth’s crust may slide or glide off regionally uplifted terranes under the influence of gravity or as the result of widespread crustal stretching. Surfaces of decoupling or detachment along which sliding takes place may involve tens of thousands of square kilometers, and the scale of folds and faults in the deformed plate above this detachment surface may be complex and measured in tens of kilometers. Growth faults, which are similarly gravity driven, can produce hazardous ruptures both at the surface and at depth. It is difficult to distinguish between those structures driven by gravity from those resulting from either shortening or stretching of the upper mantle and lower crust, and the distinction in places is moot. Other gravity-driven movements of smaller masses of Earth materials are initiated in

OCR for page 3
Active Tectonics: Studies in Geophysics regions of high relief, including, for example, landslides, debris flows, and mud flows, as well as volcanic-related lava flows, ash flows, and lahars. Even though such movements of masses result from active tectonism and may constitute severe hazards, they are considered to be secondary to the more fundamental active tectonic processes and thus are not discussed extensively in this study. RATES OF ACTIVE-TECTONIC AND GEOMORPHIC PROCESSES Even the most stable intracontinental and intraoceanic basin areas of the Earth’s surface are constantly changing, although at rates difficult to detect by classical geodetic techniques. But along the margins of continents and ocean basins where plates of the Earth’s crust collide or interact, the rates can be rapid and even catastrophic and, for volcanic eruption, explosive. Explosive rates are represented by such dramatic volcanic eruptions of historical times as Tambora (1815) and Krakatau (1883), in Indonesia, Mount St. Helens, Washington (1980), and El Chichon, Mexico (1982). Along great faults lateral displacements of about 10 m, along tens of kilometers of fault length, have occurred abruptly and suddenly to generate great earthquakes. Eyewitnesses of the 1983 Borah Peak, Idaho, earthquake reported that a meter-high scarp formed in less than a second. Laboratory and field analyses show that fault ruptures propagate at rates of 2–3 km/sec. Longer-term average rates of processes are exemplified by as much as 10 cm per year of relative motion between the major plates of the Earth’s crust. Geodetic nets a few tens of kilometers wide across the plate-bounding San Andreas Fault in California show relative motion of up to 3.5 cm/yr; this rate is confirmed close to the fault using creepmeters spanning only a few tens of meters across creeping segments of the fault. This rate contrasts with a rate of between 5.5 and 6 cm/yr for the relative plate motion between the North American and Pacific plates, as determined from the rate of seafloor spreading calculated from the positions of the magnetic stripes that lie symmetrically on the two sides of the East Pacific Rise. The difference of 2 to 2.5 cm/yr between the rate of motion on the fault and that of the plates is distributed across the western part of the North American continent, possibly affecting areas as far inland as 1000 km. Vertical displacements of the Earth’s crust can involve a sudden uplift of as much as 10 m, as occurred at Montague Island during the Alaskan earthquake of 1964. In the active regions of the Great Basin province of the western United States, average vertical displacement rates of mountain blocks relative to basin blocks not uncommonly have been as high as several tenths of a millimeter per year for several millions of years. In Japan, near the Kozo-Matsuda Fault, average displacement rates as high as 5 mm/yr over thousands of years are recorded. Although the centers of continents are generally thought to be relatively stable, average uplift rates may range from 0.01 to 1.0 mm/yr for several tens of thousands of years. Uplift rates on the Monroe Uplift, in the Mississippi Valley, for example, have been of this general rate. It is estimated that the crest of the Ventura anticline in southern California has been rising at a rate of about 10–15 mm/yr for a few hundred thousand years. Optimism about improving our capability of making meaningful short-term predictions of volcanic eruptions and developing methods to predict earthquakes is predicated primarily on improving our ability to measure and to recognize significant changes in rates of deformation. Among the signals used for successful prediction of eruptions at Mount St. Helens were increases in the rate of displacement of the surface of the new lava dome and the rate of thrusting on the crater floor adjacent to the lava dome (Figure 1). Increased rates of tilt of the flanks of Kilauea volcano in Hawaii have been recognized for several decades as indicators of magmatic filling of shallow reservoirs and impending eruptions of the volcano. Interpretations of increased seismicity—

OCR for page 3
Active Tectonics: Studies in Geophysics FIGURE 1 Prediction of eruptions. More than a dozen eruptions have been successfully predicted at Mount St. Helens, Washington. One of the important data sets for the predictions is the rate of thrusting in the floor of the crater adjacent to the lava dome. The crater formed during the explosive eruption of 1980 and the lava dome that subsequently developed are shown in this photograph of April 17, 1981. The graph shows contraction across a thrust fault on the crater floor prior to the September 1981 eruption. The black rectangle is the period within which the eruption was predicted to occur. The prediction was made at the arrow, and the eruption occurred at the vertical dashed line [graph from D.A.Swanson, T.J.Casadevall, D.Dzurisin, S.D.Malone, C.G.Newhall, and C.S.Weaver (1983), Science 221, 1369–1376; photograph by R.E.Wallace, U.S. Geological Survey]. a manifestation of strain release—have played a key role in successful predictions of volcanic eruptions in Hawaii, Washington State, and elsewhere. The concept that elastic strain must accumulate in brittle rock before rupture occurs is the underlying rationale for earthquake prediction. Changes in the rate and patterns of strain just before rupture that are seen in laboratory experiments suggest that marked changes—either decreases or increases in rates of strain or in sense of strain—may signal an impending rupture and earthquake. A 2-yr halt in fault creep at one point on the Calaveras Fault prior to the Coyote Lake earthquake of 1979 in central California may have been a precursor to that earthquake. Regional tilts of land surfaces in Japan before the 1964 Niigata earthquake may represent the type of strain to be expected, but questions have been raised about older surveys and their adequacy for accurately recording such changes. However, few, if any, recognized changes that have been documented are widely acknowledged as having been clear premonitory signals for earthquakes. Tectonic rates are reflected in rates and patterns of geomorphic processes and forms. For example, fault-generated range fronts on which displacement rates average tenths of a millimeter per year for a million or more years are characterized by low sinuosity of the topographic boundary between the range and basin and by faceted spurs. The range margin sinuosity increases with slower rates, and the pattern of faceted spurs gives way to more highly dissected and less planar landforms along range fronts. The rates of geomorphic processes range dramatically from climate to climate, and different processes may be dominant in each. For example, fluvial processes are dominant in semiarid areas, whereas wind action may be dominant in arid regions and mass movement in some humid or tropical regions. In humid climates the potential for sediment production could be great, but vegetation reduces erosion rates. These and other

OCR for page 3
Active Tectonics: Studies in Geophysics processes of erosion, such as aeolian, raindrop impact, ice, and geochemical processes, interact in different climates to affect collectively the overall degradation and destruction of landforms constructed by tectonic processes. Slopes of fault scarps or wave-cut cliffs may be modified rapidly by gravitational processes. The near-vertical free face of a few-meter-high scarp may disappear within a few hundred years, as material falls off the steep face to rest at the angle of repose in a debris slope covering the free face. The scarp later becomes modified by downslope movement of material according to a diffusion model, which operates at slower and slower rates. However in a desert climate, a meter-high fault scarp may be clearly recognizable for a hundred thousand years. TECHNIQUES FOR EVALUATING ACTIVE TECTONICS The entire range of geologic, geophysical, and geodetic techniques may have a bearing on evaluating active tectonics, but for studying ongoing processes some are more useful than others. The period of past behavior of tectonic movements that is significant to predict future movements may range from days to thousands, or even millions, of years. In general, however, the predictive value of events of the past decreases with age, but the length of time analyzed must be sufficiently long to sample adequately a particular series of events, changes in rates of events, or changes in patterns of tectonics. For example, large intraplate earthquakes generally recur on a given fault at intervals of several thousands of years, and, thus, an estimate of earthquake potential based on only a century or two of recorded history may mistakenly suggest quiescence in regions that hold the most severe threat of great earthquakes. Each sample interval of time provides different insight into tectonic processes. Repeated geodetic measurements made over days or decades provide details of tectonism unidentifiable in most geologic studies, but only from the longer-term geologic record can many tectonic movements be identified and their rates determined. In regard to active tectonics, the segment of the future of greatest concern to mankind generally is only the next few years to few decades, although for safe disposal of some radioactive waste the period is thousands or tens of thousands of years. Events of the past 100,000 to 200,000 yr, and especially of the past 10,000 to 20,000 yr, are particularly significant as a basis for predicting future trends. Geodetic techniques can be used to identify and quantify very recent historical tectonism, and real-time geology and geophysics have already been used to provide predictions of specific future events. Some of the greatest successes have been achieved in predicting eruptions of volcanoes. The basic physical and chemical models that can be translated into geophysical, geochemical, and geologic predictions are in the process of rapid evolution, and many advances seem to be just over the horizon. Though this emphasis is on events of late Quaternary time, we do not exclude significant evidence about tectonic history and processes obtainable from longer periods of geologic time, for example, the past few million years, or latest Neogene and Quaternary time. Neotectonics, which pertains to the tectonics of this longer period of time, is the focus of the Neotectonic Map Project under the Geological Society of America’s Decade of North American Geology (DNAG) project. For this review, neotectonics is distinct from active tectonics but is recognized as providing an important set of data. Even the longer record cannot be ignored, because to a large extent those structures that originated in Mesozoic and Tertiary time greatly influence the patterns of Quaternary structures. Paleozoic and older structures have less influence. Stress orientations may have changed, but during any period of strain the prefractured nature of the Earth’s crust and other anisotropism affect the response of the crust. Furthermore, the tectonics of old features now inactive can cast light on those currently active.

OCR for page 3
Active Tectonics: Studies in Geophysics A strategy for geodetic monitoring of active tectonics was concisely presented in Geodetic Monitoring of Tectonic Deformation—Toward a Strategy (1981), and a few highlights of that study serve to indicate the suggested types of problems, monitoring techniques, and evaluation procedures. Questions exist, for example, about the relationship between the Chandler wobble of the Earth and earthquakes and whether major earthquakes cause a discernible change in the polar path. Up to a point, classical astronomical observations can be used to study such problems. However, improved geodetic space techniques such as very-long-baseline radio interferometry and satellite laser ranging are rapidly advancing our understanding of plate motions; such observational techniques are beginning to allow the direct measurement of present-day plate motion. Such studies can cast light on the rheological properties of the asthenosphere, its viscosity, and whether gross strain is rather more impulsive than continuous. Regional strain measurements are currently made by laser-ranging techniques in which measurements are infrequent. By employing two-color laser-ranging techniques, corrections for atmospheric conditions are made automatically in data processing and very frequent observations are practical. Small trilateration nets, level lines, and stretched-wire creepmeters spanning only a few tens of meters are used for studying localized deformation along faults. Tiltmeters, linear strain meters, and volumetric strain meters are among instruments currently used to study tectonic strain. The precision required is of the order of a few parts in ten million, but some measurements of the order of a few parts in a billion or smaller are desirable. Most instruments do not have, or barely have, such capabilities. Noise near the Earth’s surface, furthermore, exacerbates the problem of recognizing tectonic signals of such small size. Long-term stability of instruments is a special problem. The historical record prior to modern instrumentation can provide important insight into longer-term rates and patterns of volcanism, earthquakes, and crustal deformation. In an area of active uplift in Iran, for example, a canal built 1700 yr ago has cut down about 5 m below its original bed, indicating an uplift rate of about 2 mm/yr. In western North American, the historical record of earthquakes is not even 200 years long, but in China it is almost 4000 yr long, although reasonably complete for only about 1000 yr. Similarly there is a historical record of volcanism that is useful as a measure of the frequency and characteristics of eruptions. Because of low rates of tectonic strain accumulation in many places, the periods of time represented even by recorded history do not adequately sample longer-term trends and cycles. The historically most inactive sections of great faults, indeed, may be candidates for generating the largest future earthquakes. Geologic techniques provide the only means to sample sufficiently long time spans during which many rates and patterns of tectonic processes must be analyzed. The stratigraphic record contains clear physical evidence of past events. In Japan, for example, bountiful evidence of complex volcanism is preserved by layer upon layer of volcanic ash, breccias, and lava flows. Furthermore, abundant tephra layers—each of which has a distinctive chemical and mineralogic characteristic and was deposited within a few days or weeks—over large areas provide distinctive time markers that permit many local and regional tectonic events to be analyzed in the context of a time frame. Stratigraphic relations along active faults may reveal a sequence of successive offsets and deposition of sediments that permit the reconstruction of the history of faulting, the recurrence intervals, and size of faulting events. Sedimentary structures such as sand blows, clastic dikes, and deformed beds related to liquefaction may reveal the history of strong ground shaking and add to the evidence of prehistoric earthquakes. Repeated deposition of colluvial wedges along the bases of fault scarps creates unique stratigraphic relations from which an interpretation of the paleoseismological record can be derived.

OCR for page 3
Active Tectonics: Studies in Geophysics Analysis of gross stratigraphy—the composition of the lithologic units and the deformation of beds—can indicate whether rates of uplift nearby were rapid or slow, whether volcanic arcs were nearby, or whether terranes that contributed clasts to the sediments had later moved away. At small scales, as in excavations dug across active faults, the sequence of sedimentation and faulting can be analyzed to determine how often large prehistoric earthquakes occurred. The patterns and rates of silting of rivers are extremely sensitive to tectonic changes. Landforms, created by the competition between tectonic constructional and erosional destructional processes, contain abundant evidence of active tectonics. Classical geomorphology, however, has concentrated primarily on descriptions of landforms, and process and stages of erosion were stated in only generalized terms. In the past decade or so, quantitative, process-oriented geomorphology has been directed toward problems of active tectonics. Studies of marine terraces, river geomorphology, fault scarps, and eruptive volcanic features have demonstrated a rich and readily available source of information about active tectonics recorded in landforms. Coseismically uplifted marine terraces may provide evidence of prehistoric earthquakes. Direct measurement of fault-scarp morphology, for example, when analyzed according to an error-function solution of a diffusion model, can approximate the date of a great prehistoric earthquake. From such analysis, the pattern and recurrence interval of large earthquakes can be determined even where average recurrence intervals are measured in thousands of years. In summary, as more research is carried out on the effect of active tectonics on landforms, geomorphology can become one of the most powerful tools for evaluating active tectonics. Clearly, tectonic processes are complex, often nonuniform, and the emphasis given here on predicting future tectonic activity introduces the important problem of how best to state the likelihood of future events. A variety of formal mathematical methods is currently available for expressing probability, but further study is warranted into techniques of conveying to both lay and technical audiences the likelihood and consequences of complex tectonic activity. Applications of such techniques as described above have recently led to some new insights about active tectonics. The findings described below, furthermore, indicate some areas of research that are worthy of further attention. Individual or groups of large prehistoric earthquakes can be clearly identified by geologic means, such as by microstratigraphy and microgeomorphology. The research efforts so directed are now termed paleoseismology. By such methods the long-term patterns and timing of some great earthquakes have been analyzed. Coherence of signals among gravitational changes, lateral changes, and vertical changes confirm that tectonic changes on a time scale of months or years are real and are amenable to analysis. From such studies the normal dynamic behavior of the Earth’s crust may be determined, and the predictive value of unusual changes can be assessed. A variety of active-tectonic realms of diverse sizes has been recognized and defined; each realm is characterized by distinctive patterns and rates of deformation. Such analysis provides a basis for zonation to assist in the reduction of geologic hazards. Folding can be an important active-tectonic process. In the past, folding has been ignored to a large extent in active-tectonic studies even though a rich and voluminous mass of information about folding has been in existence for a long time in the classical tectonic literature. Further analysis will permit evaluation of the strain budget, that is, the distribution of strain, between faulting and folding. Some well-known physical models such as the diffusion or heat-flow model have been applied successfully to the analysis of geomorphic processes. Determining the distribution of microseismicity has proven to be one of the best

OCR for page 3
Active Tectonics: Studies in Geophysics techniques for defining the pattern of active faulting. Such studies can “paint” the three-dimensional time-space relation and show how strain is propagated. Additionally, digitally recorded data from broadband, wide dynamic-range seismometers have permitted greatly improved analysis of the faulting process. Broad vertical changes in elevation that occur on an historical time scale, within what are generally thought of as stable intracontinental regions, have been found to affect dramatically the regimes of major rivers, bank stability and silting, and navigation and flood control. IMPACT ON SOCIETY Only when an infrequent paroxysm of the Earth’s crust occurs in the form of an earthquake, volcanic eruption, or landslide is the dynamic nature of the Earth’s crust brought clearly to public attention. Largely unnoticed are slower movements of the Earth’s crust, which, nevertheless, are costly in terms of engineering countermeasures required or the constraints that are placed on land uses. The demands for safety and continuity of service required of modern-day critical facilities, such as nuclear reactors, large dams, and structures for defense, require a knowledge of the Earth’s crust well beyond the current state of geophysical and geologic information. As a result, costly mistakes have been made, and high-priority programs have been delayed or canceled. Without a doubt the future will call for more such uses in projects having even greater sophistication, greater potential hazard, and more complex interactions with the environment. The Tragedy of Ignorance For the want of information about active tectonics, numerous mistakes and extremely costly delays or cancellations of major engineering projects have resulted. A few case histories will serve as examples. In the early 1960s, the Pacific Gas and Electric Company began to develop a site for a nuclear reactor at Bodega Bay in northern California. The first site selected was astride the part of the San Andreas Fault that caused the great San Francisco earthquake of 1906. Soon, however, it became recognized that the fault might pose problems, and the site was moved a few kilometers west, but still near the fault zone. During the excavation of the giant pit (Figure 2) that was to contain the reactor, faults were found and concern was raised that movement on these secondary faults might rupture the reactor and cause a major disaster. At that time, and to a major extent even today, the geologic and geophysical communities simply had no scientifically based answers to such fundamental questions as: Will the next rupture recur on the 1906 break of the San Andreas Fault, or will it occur on some other fault branch or strand within the kilometer-wide fault zone? How much displacement can be expected on branch faults and parallel faults at distances of several kilometers from the main break? How does a history of one displacement in 10,000 or 35,000 yr on a branch fault factor into a calculation of probability of a future displacement on that branch fault? In 1969 the Bodega Bay nuclear reactor project was canceled after several years of acrimonious debate, because no one had the needed scientific information with which to assess the stability of the site. As a result millions of dollars were spent unproductively. The lack of understanding of active tectonics has caused problems in the development of similar facilities. As an additional example, in 1979 the building of a large thin-arch dam at Auburn, California, in the foothills of the Sierra Nevada was stopped. Although much had been learned about earthquake hazards in the ensuing years since the Bodega Bay nuclear

OCR for page 3
Active Tectonics: Studies in Geophysics FIGURE 2 Wasted effort! Excavation at Bodega Bay, California, for a proposed nuclear reactor began in the early 1960s. The San Andreas Fault lies under Bodega Bay in the background, and branch faults were found in the excavation. Concern over faulting events that might rupture the reactor caused cancellation of the project. Because of the infancy of the science of faulting and relation to earthquakes, scientists could not agree on the probability of a faulting event or the size of such an event under the site (photograph by R.E. Wallace, U.S. Geological Survey). reactor project was canceled, data about the earthquake and fault-displacement potential of the Sierra Nevada foothills belt were almost nonexistent. A team of earth-science consultants studied the problem for the U.S. Bureau of Reclamation and estimated that several inches of displacement along the fault was very possible within the lifetime of the dam. Government geologists, serving in a review capacity, estimated that 3 feet of displacement might occur on the faults during an earthquake. Clearly, there was no satisfactory scientific basis for agreement on the exact amount of displacement to be expected, but both groups of specialists did agree that fault displacement during the lifetime of the dam must be considered a distinct possibility. After review of the fault data and the engineering characteristics of the dam, officials of the U.S. Bureau of Reclamation concluded that even the smaller estimated displacement would exceed the accommodation capabilities of the dam as planned. After 10 yr of preliminary site development at a cost of over $200 million, the project was halted (Figure 3). Benefits from Understanding The veil of ignorance, at least with regard to earthquake hazards in the United States, has been rolled back to some extent as a result of the National Earthquake Hazard Reduction Program, which formally began in 1977. Fundamental studies of volcanic processes at the Hawaiian Volcano Observatory, similarly, have resulted in practical guidelines for reducing the hazards of volcanic eruption. Examples of successes can serve to illustrate the kinds of benefits that are likely to accrue from long-term fundamental research in active tectonics. On October 28, 1983, an earthquake of magnitude 7.3 occurred in central Idaho. A spectacular fault scarp 36 km long formed along the west flank of the Lost River Range, a typical Basin and Range fault-block mountain (Figure 4). Significantly the 1983 rupture occurred exactly where it was expected in that it replicated a fault displacement of similar size that occurred in late Holocene time, possibly within the last 5000 yr. Only in the past decade had the tendency been demonstrated

OCR for page 3
Active Tectonics: Studies in Geophysics clearly that historical breaks occur along the traces of Holocene breaks. The fault along the Lost River Range had been studied in 1977 to evaluate the earthquake hazards at the Idaho National Engineering Laboratory (INEL), and Holocene displacement had been identified. As a result of correctly anticipating the location and size of the 1983 earthquake, automatic shut-off systems functioned effectively at INEL and closed down nuclear reactors, and a potentially hazardous situation was countered. An impact on policy and decision making that can result when a significant new set of data and some basic understanding about active tectonics replace almost total ignorance can be illustrated by discoveries of the late 1970s and early 1980s in the Mississippi Valley. There, until about 1981–1982, the causative structures of the great New Madrid earthquakes of 1811–1812 were mysteries. As a result, in planning an engineering project, the earthquakes were considered random events that could occur anywhere over a vast area of the midcontinent. A combination of geophysical explorations including FIGURE 3 Canceled! After years of site preparation the thin-arch dam planned near Auburn, California, was canceled because faults of the Foothills Fault zone were found to cut the rocks under the foundation. Analyses indicated the possibility of displacements too large to be accommodated by the planned design of the dam. Almost nothing was known about the Foothills Fault zone before the dam was started (photograph courtesy of the U.S. Bureau of Reclamation).

OCR for page 3
Active Tectonics: Studies in Geophysics FIGURE 4 Faulting site predicted. The vertical bare slope is the fault scarp that formed during the 1983 earthquake at Borah Peak, Idaho. An older, degraded scarp is represented by the more gradual slope between the man’s feet and the top of the vertical scarp. Analysis of the slope of the older scarp and microstratigraphy exposed in 1977 in a trench through this site indicated previous displacement on the fault within the past several thousand years (Holocene time). Recently gained understanding of fault behavior strongly indicated that such Holocene fault scarps are candidate sites for future displacement, and this theory was confirmed here. As a result of the studies, nuclear reactors at the nearby Idaho National Engineering Laboratory were equipped with automatic shut-off systems that functioned successfully during the earthquake (photograph by R.E.Wallace, U.S. Geological Survey). gravity, magnetic, and seismic reflection techniques revealed a buried but clearly definable rift zone (Figure 5). Faults of the rift zone were shown to be ancient geologic features that originated as extensional features but that were being reactivated by compressional stresses in the current tectonic regime. Geologic studies suggested a recurrence period of 600–1000 yr for such great earthquakes. Although much is yet to be learned, the structures of the rift zone that caused the 1811–1812 earthquakes, and which will likely generate future earthquakes, now are clearly recognizable. Thus, potential future earthquakes to be accounted for in the design of engineering projects need not be considered possible everywhere in the mid-continent; engineering designs can be much more realistic and cost-effective, and regional planning can be conceived on a more rational basis. Much of the central part of the United States has been considered to be free of major earthquake hazards. Surprisingly, however, in a seismically very quiet part of central Oklahoma, the Meers Fault has recently been discovered to have had displacement of several meters in Holocene time, presumably accompanied by one or more major earthquakes [M.C.Gilbert, Earthquake Notes 55(1), 1–3 (1985)]. A remarkably fresh fault scarp is still preserved, and the last event may have occurred within the past few thousand years. The discovery raises the question as to how many other unidentified active faults exist in seismically quiet regions of intracontinental regions. Newly developed methods in paleoseismology now permit investigation of this question. Socioeconomic Measures Measures to counter the effects of earthquakes, volcanic eruptions, and slower forms of tectonic deformation fall into four categories: engineering accommodations, sensible

OCR for page 3
Active Tectonics: Studies in Geophysics FIGURE 5 Causative structures revealed. Until various geophysical techniques were brought to bear in the search for possible buried faults that could have caused the great 1811–1812 earthquakes, similar earthquakes were presumed to be possible over a vast area of the Mississippi Valley. As a result, concern was raised about major engineering projects. This map of the second vertical derivation of the total magnetic-field intensity reveals a northeast-trending magnetic feature that is interpreted as a completely masked graben. Historical seismicity has been concentrated within the graben, and thus the rift structure is presumed to be responsible for the generation or control of seismicity. [Figure from T.G.Hildenbrand, M.F.Kane, and J.D.Hendricks (1982), Magnetic basement in the upper Mississippi Embayment region—A preliminary report, in Investigations of the New Madrid, Missouri, Earthquake Region, U.S. Geol. Surv. Prof. Paper 1236-E, pp. 38–53.] use of the land, event prediction, and disaster preparedness. Engineering design and construction of structures to resist earthquake shaking, offset by faulting, and differential settlement can greatly reduce the hazard of structural collapse. Planning the use of land so that structures, especially those for dense occupancy, are not built astride active faults or in the potential paths of volcanic lava flows or lahars is prudent and effective, and yet such obvious methods of reducing hazards commonly have been ignored. In many states and countries, building codes and standards are followed to some extent, but in practice the craftsman or builder, through ignorance or neglect, too often fails to incorporate the specified design features in the final structure. Simple factors such as the lime content of mortar and the proper wetting of bricks before application of mortar in masonry construction greatly affect the strength of buildings. Special studies are required when building near active faults in California. Disaster plans for responding to potential hazards from volcanic eruptions have been prepared in California, Washington, Japan, and Italy, among others. Successful prediction of at least a few dozen volcanic eruptions undoubtedly has reduced the hazard to life and property. Among damaging earthquakes for which claims of prediction have been made, only the Haicheng, China, earthquake of 1974 seems to be widely credited by the scientific community as having had a valid scientific basis. For the first time in the United States, a long-term earthquake prediction was reviewed and approved as scientifically valid by both the National Earthquake Prediction Evaluation Council and the California Earthquake Prediction Evaluation Council. The prediction was announced officially by the Director, U.S. Geological Survey, on April 5, 1985, and the announcement stated that “an earthquake of magnitude 5.5 to 6 is likely to occur in the Parkfield, California, area within the next several years (1985–1993)….” Identification of the most stable tectonic blocks and prediction of the degree of stability, or lack of active tectonism, have been even more difficult than predicting tectonic events. Few methods of analysis and little understanding of the long-term processes are at hand. The past two decades have seen a major acceleration in disaster preparedness. One example is the Southern California Earthquake Preparedness Project, which was

OCR for page 3
Active Tectonics: Studies in Geophysics started in 1980 under joint state and federal support. Preliminary prototype response plans have been completed for an industry, a county, a large city, and a small city. Elements of the prototype plan include, for example, voluntary self-help, communications, special problems of schools, law enforcement, and concerns of financial institutions. An early requirement for the project was an accurate scenario of the tectonic process, in this case the detailed effects of a major earthquake when it happened and the likely content of an earthquake prediction if it were made. These needs could be fulfilled only poorly or not at all because of the state of earthquake science. All administrative, societal, economic, and other steps to cope with the impact of active-tectonic processes hinge on understanding those processes and the rates at which they operate. RECOMMENDATIONS FOR RESEARCH PRIORITIES AND ACTIONS A program of fundamental research focusing especially on Quaternary tectonic geology and geomorphology, paleoseismology, neotectonics, and geodesy is recommended to better understand ongoing, active-tectonic processes. Capability should be developed to assess the potential for, or to predict, tectonic activities up to several thousand years in the future. To accomplish these goals, special attention should be given to the following research areas that constitute especially powerful techniques or gaps or areas of weakness within the complex matrix of scientific disciplines needed. We stress strongly, however, that the priorities given here are predicated on the existence of a healthy level of fundamental research in related, broad subjects including understanding the continental lithosphere, plate tectonics, energetics of tectonic processes, seismology, and volcanism. Other National Research Council reports have already addressed many of these topics. Research Priorities Dating Techniques: The greatest need is for data and models concerning the rates of active-tectonic processes. Public-policy decisions, for example, about what levels of earth-hazard risks are acceptable, rest largely on evaluations of the rates (and the variability of rates) of processes, the frequency of events, and the prediction of when hazards might become critical. To this end, new and improved techniques of dating materials and events of Quaternary age are needed. For example, thermoluminescence, uranium trend, beryllium-10, aluminum-26, and soil development are among two dozen or more potentially useful dating techniques currently being investigated as alternatives to the more widely used techniques, including radiocarbon dating. Tandem accelerator mass spectrometry may permit several of these new techniques to become practical and may improve the usefulness of older techniques. The breadth of applicability and precision of each technique need improvement. Tectonic Geomorphology: Research in tectonic geomorphology aimed at documenting rates, styles, and patterns of movements is recommended as one of the potentially most effective means of analyzing ongoing tectonic processes. Research in quantitative geomorphology is especially important. Geodesy: Research using geodetic techniques should be expanded to delineate the patterns and scales of ongoing deformation. We endorse the recommendations in the 1981 report Geodetic Monitoring of Tectonic Deformation—Toward a Strategy by the Committee on Geodesy/Committee on Seismology. Techniques described in that report range from ground techniques using strain meters and laser-ranging devices to space techniques using very-long-baseline radio interferometry (VLBI) and the Global Positioning System (GPS). Paleoseismology: Because major tectonic processes such as those manifested by seismicity are commonly grossly misrepresented by the historical time sample, pa-

OCR for page 3
Active Tectonics: Studies in Geophysics leoseismic techniques should be furthered. Physical exploration, such as trenching to expose and permit microstratigraphic analysis in critical places, and geomorphic techniques aimed at paleoseismology are of high priority. Real-Time Geology: Recording, processing, and interpreting of geologic changes in a time frame of seconds to weeks—which can be termed “real-time geology”—are needed for the development of short-term predictive capability. Special instrumentation, automation, and rapid computer processing of data are required. A far more detailed understanding than currently exists of certain real-time Earth processes is needed. Probability Studies: Methods for expressing the probability of future tectonic activity need much improvement. Incomplete data and debatable interpretations of data will continue to be a problem, and yet both conclusions and uncertainties must be communicated as precisely as possible within the technical community as well as to nontechnical decision makers. Recommendations for Specific Actions In order to advance these research areas of high priority, the following institutional considerations and technical approaches are emphasized: Research in dating techniques suitable for analysis of Quaternary geology should be (a) recognized as a high priority for support by funding groups—both governmental and private and (b) encouraged through workshops and symposia under the auspices of professional societies. Studies of Quaternary geology useful in analyzing active tectonics should be more effectively integrated into the earth-science curricula of universities. Special attention should be given to tectonic geomorphology including the observational, experimental, and quantitative-theoretical elements. Emphasis should be given to research programs that employ geodesy to analyze the dynamic as well as the static features of the Earth’s crust. Clear and specific identity should be given to research concerning Quaternary structures, processes, and rates of processes. Research programs to develop short-term (up to a few weeks) predictive capabilities for active tectonics, along with the ability for rapid data analysis, should be established or accelerated by those federal and state agencies responsible for issuing geologic hazard warnings. To this end, regional seismic networks and systems for monitoring strain and stress in real time are critical. Preplanning is needed to assure the effective study of events of opportunity. The observation of volcanic eruptions, earthquakes, or landslides while they are occurring can reveal facts about Earth processes unavailable at any other time. For example, recording of strong ground motion during an earthquake, measurement of changes in pore pressure during a landslide, or the changes in composition of gases during volcanic eruptions can be carried out only at specific critical moments. The continuity of carefully selected programs aimed at gathering long-term baseline data with temporal significance is desirable. The importance of regional seismic networks to the study of active tectonics should be considered and should continue to be evaluated by the appropriate institutions. Wider use of trenches, tunnels, drill holes, and other artificial excavations to reveal structural and age relations of tectonic units is desirable. Instrumentation of tunnels and boreholes, furthermore, can avoid spurious signals (noise) found near the Earth’s surface and thus permit recording of smaller signals significant to active tectonics. The further and more complete use of aerial photography and remote sensing at

OCR for page 3
Active Tectonics: Studies in Geophysics many scales and in many radiation bands can reveal or clarify geomorphic and geologic relations otherwise unavailable. Side-looking radar, which permits illumination of the landscape from angles impossible by natural sunlight, can enhance selected geomorphic and tectonic trends. New, sophisticated technology, such as used in the airborne profiling of terrain (APT) system, may provide previously unobtainable data on geodetic and geophysical changes that are a direct reflection of active tectonics. BIBLIOGRAPHY U.S. Program for the Geodynamics Project: Scope and Objectives, U.S. Geodynamics Committee, National Research Council, National Academy of Sciences, Washington, D.C., 235 pp., 1973. Predicting Earthquakes, Committee on Seismology, National Research Council, National Academy of Sciences, Washington, D.C., 62 pp., 1976. Trends and Opportunities in Seismology, Committee on Seismology, National Research Council, National Academy of Sciences, Washington, D.C., 158 pp., 1977. Earthquake Hazards Reduction: Issues for an Implementation Plan, Office of Science and Technology Policy Executive Office of the President, Washington, D.C., 230 pp., 1978. Geophysical Predictions, Geophysics Study Committee, National Research Council, National Academy of Sciences, Washington, D.C., 215 pp., 1978. An Assessment of the Consequences and Preparations for a Catastrophic California Earthquake: Findings and Actions Taken, Federal Emergency Management Agency, Washington, D.C., 59 pp., 1980. Continental Tectonics, Geophysics Study Committee, National Research Council, National Academy of Sciences, Washington, D.C., 197 pp., 1980. Earthquake Research for the Safer Siting of Critical Facilities, Committee on Seismology, National Research Council, National Academy of Sciences, Washington, D.C., 49 pp., 1980. Dynamics and Evolution of the Lithosphere: The Framework for Earth Resources and the Reduction of Hazards, Inter-Union Commission on the Lithosphere, International Council of Scientific Unions, 62 pp., 1981. Geodetic Monitoring of Tectonic Deformation—Toward a Strategy, Committee on Geodesy/Committee on Seismology, National Research Council, National Academy Press, Washington, D.C., 109 pp., 1981. Goals and Tasks of the Landslide Part of a Ground-Failure Hazards Reduction Program, U.S. Geological Survey, 49 pp., 1982. Effective Use of Earthquake Data, Committee on Seismology, National Resource Council, National Academy Press, Washington, D.C., 51 pp., 1983. The Lithosphere, Report of a Workshop, U.S. Geodynamics Committee, National Research Council, National Academy Press, Washington, D.C., 84 pp., 1983. Multidisciplinary Use of the Very Long Baseline Array, Board on Physics and Astronomy, National Research Council, National Academy Press, Washington, D.C., 202 pp., 1983. Opportunities for Research in the Geological Sciences, Board on Earth Sciences, National Research Council, National Academy Press, Washington, D.C., 95 pp., 1983. Seismographic Networks: Problems and Outlook for the 1980s, Committee on Seismology, National Research Council, National Academy Press, Washington, D.C., 62 pp., 1983. Explosive Volcanism: Inception, Evolution, and Hazards, Geophysics Study Committee, National Research Council, National Academy Press, Washington, D.C., 176 pp., 1984. National Earthquake Hazards Reduction Program: Overview, M.L.Schnell and D.G.Heard, eds., U.S. Geological Survey Circular 918, 65 pp., 1984. Seismological Studies of the Continental Lithosphere, Committee on Seismology, National Research Council, National Academy Press, Washington, D.C., 144 pp., 1984. Geodesy: A Look to the Future, Committee on Geodesy, National Research Council, National Academy Press, Washington, D.C., 179 pp., 1985. Safety of Dams: Flood and Earthquake Criteria, Water Science and Technology Board, National Research Council, National Academy Press, Washington, D.C., 321 pp., 1985.