USArray: A synoptic investigation of the structure, dynamics, and evolution of the North American continent

 

 

During 1999, two important workshops, one in Albuquerque NM, and a second in Houston TX, provided a forum to discuss the design and implementation of an ambitious plan to develop an integrated understanding of North American geology and deep Earth structure, the USArray Initiative. These workshops jointly sponsored by the National Science Foundation and IRIS (Incorporated Research Institutes in Seismology) were attended by a broad spectrum of Earth scientists with representatives from academia, the US Geological Survey, regional networks, and the National Science Foundation. Workshop participants identified scientific challenges, the USArray initiative helped define the technical and multidisciplinary components of the USArray facility, discussed an operation and management scheme for the facility and identified ways in which USArray can best be used to advance Earth science research, education and outreach.

This white paper and the attached appendices, prepared by the USArray Steering Committee, document the outcome of these two workshops.

The USArray Steering Committee includes:

Anne Meltzer (Coordinator), Lehigh University

Rick Carlson, Carnegie Institute of Washington

Tim Dixon, Miami University

Göran Ekström, Harvard University

Michael Gurnis, California Institute of Technology

Gene Humphreys, University of Oregon

Karl Karlstrom, University of New Mexico

Alan Levander, Rice University

Roberta Rudnick, Harvard University

Peter Shearer, University of California, San Diego

Paul Silver, Carnegie Institute of Washington

Rob van der Hilst, Massachusetts Institute of Technology

Peter Zeitler, Lehigh University.

 

Additional input to the appendices comes from Gabi Laske, Guy Masters, Frank Vernon, and John Orcutt (University of California, San Diego), Steve Park (University of California, Davis), Catherine Johnson (IRIS), Randy Keller (University of Texas, El Paso), and Robert Harris (University of Utah).

 

USArray: A synoptic investigation of the structure, dynamics, and evolution of the North American continent

White Paper Contents:

1. Executive Summary

2. Rationale for the facility

3. Development of the initiative

4. Outline of the facility

4.1 The Geophysical Component of USArray

4.2 The Geologic Component of USArray

4.3 Earth Science Information System

4.4 Education and Outreach

4.5 Operational Mode, Community Input, Science Support, Partnerships

4.6 Beyond Research

5. Summary

6. Recommendations

Appendices:

Appendix 1: The Seismic Component of USArray

1.1 History of the USArray Initiative

1.2 USArray Imaging Science Technical Specification

1.2.1 Earthquake Detection and Frequency

1.2.2 Resolution Measures for Direct Wavefield Imaging

1.2.3 Direct Teleseismic Imaging with USArray

1.2.4 Teleseismic Imaging using Surface Waves

1.2.5 Active Source and Microearthquake Seismology

1.3 Development of a Virtual Seismic Network

1.4 Station Operations within USArray Appendix 2: USArray and the Deep Earth

Appendix 3: The Geologic Component of USArray

Appendix 4: The GPS Component of USArray

Appendix 5: The Magnetotelluric Component of USArray

Appendix 6: Education and Outreach

Appendix 7: USArray and Continental Margins

Appendix 8: Synergy with other EarthScope Initiatives

Appendix 9: Viewgraphs/Overheads

Attachments:

IRIS Newsletter article

EOS Article

GSA Today Article

Science Article (Fischer and van der Hilst)

Workshop (USArray I and II) Overviews/Programs/Participant Lists

 

USArray: A synoptic investigation of the structure, dynamics, and evolution of the North American continent.

1. Executive Summary. Images profoundly shape our view of the world. Twenty-five years ago, the simple image of Earth from space brought a fractured world new attitudes about the fragile beauty and finite limits of our home planet. Today, we treat such views of Earth as commonplace but the impact of these images is no less profound. In a slightly different vein, satellite images routinely track weather and storms providing critical information to mitigate weather-related hazards. Courtesy of television news programs, people in all walks of life can now appreciate and visualize the complex dance of atmospheric dynamics to a degree previously unthinkable. Without question, synoptic views of Earth systems inevitably revolutionize our comprehension of our environment and the way we react to our natural surroundings.

USArray offers an opportunity to extent the imaging revolution to the solid Earth by developing striking new images of the North American continent, images that will, if experience is any guide, change our views of continental dynamics, plate tectonics, and our relationship to our physical environment. Plate tectonics, as developed over the last 30 to 40 years provides a framework that explains many first-order observations such as the global distribution of earthquakes, volcanoes, and mountain belts, and the relative heights and ages of continents and ocean basins. Plate tectonics works well explaining the structure and dynamics of oceanic plates, but does not provide an encompassing theory of the continents. Continental lithosphere consists of complex assemblages of crustal and mantle rocks that exhibit a far greater degree of lateral and vertical heterogeneity than their oceanic counterparts. The continental architecture, developed over billions of years, preserves the record of 95% of Earth history. Recent studies utilizing high-resolution seismic images and potential field, geochemical, and geologic data indicate there are important processes involved in continental dynamics that are only now being revealed and that are as yet not understood. Unlike the oceanic lithosphere, the long lived inherited structures and compositional variations in the continental lithosphere modulate modern tectonic processes and deformation, and likely play a crucial role in determining the patterns of the global convection system. While continents make up less than one third of the surface area of the Earth, they form the bedrock on which we live, and the continents and their margins provide all of our petroleum and most of our mineral resources. Feedbacks between tectonic and surficial processes shape the surface and crustal environment where we build our cities, get our water and other natural resources, and cope with geological hazards.

USArray also offers an opportunity for a new integrated approach to the way we conduct Earth science experiments in North America. It is time to bring together the remarkable advances made in many fields of the geological sciences to address in detail some of the most outstanding problems in Earth structure and dynamics: namely, the physical and chemical processes of continental growth and restructuring through tectonic, magmatic, and erosional processes. This includes a range of problems from studies of how a continent is associated with the underlying upper and lower mantle, to how plate margins are constructed and deform, to how near-surface mass flux (chemically and physically mobilized) couples to tectonics below and hydrologic processes above, to the structures that enable the plate to violently shift in earthquakes. While all of the continents are unique, there are obvious scientific, political, societal, financial, and educational reasons to begin this project in the United States. The US is ideally located with respect to global seismicity to provide unprecedented views of the very deep mantle. The rich fabric of tectonic provinces in North America provides a solid scientific rationale for a major national program to investigate in detail the relationships among processes and structures over a wide range of scales within the crust, lithosphere, and mantle, with the goal of understanding the tectonic and geologic processes that have made and continue to make continents.

The core of USArray is a continental scale seismic observatory designed to provide a structured yet flexible foundation for integrated studies of the continental lithosphere and deep Earth structure over a wide range of scales. The USArray initiative integrates a dramatic improvement in the resolution of seismic images of the continental lithosphere and deeper Earth with a diversity of geological data to address significant unresolved issues of continental structure, evolution, and dynamics. As an observational tool, USArray can provide new insight and new data to address fundamental questions in earthquake physics, volcanic processes, core-mantle interactions, active deformation and tectonics, continental structure and evolution, geodynamics, and fluids (magmatic, hydrothermal, and meteoric) in the crust. The USArray facility will consist of three major seismic components: (1) a transportable array of 400 broadband seismometers that will systematically cover the US, (2) ~2400 seismometers of various types (the "flexible array") designed to augment the transportable array so that a range of specific targets can be addressed in a focused manner, and (3) a permanent network of high-quality seismic stations coordinated with the US Geological Survey within the context of the National Seismic Network. The goal of this layered design is to achieve imaging capabilities that flexibly span the continuous range of scales from whole Earth, through lithospheric and crustal, to local.

USArray can also have a major impact on changing the way Earth scientists conduct experiments and convey their results to the public. The infrastructure provided by the USArray seismic facility presents a platform for a multidisciplinary field laboratory that can, over the next decade, fundamentally change the way Earth scientists carry out their investigations. USArray can offer new mechanisms for collaboration, data integration, and data management of a diverse suite of geologic, geochemical, and geophysical data sets. This will provide the catalyst to begin to harness our expanding database and to combine diverse perspectives into new synoptic images. The geologic component of USArray will provide the time dimension in understanding how continents evolve and the processes that shape them. The USArray education and outreach component will provide a new centralized framework for Earth science education at all levels that is as important and encompassing in its scope as USArray will be for scientific advances. The outcome will be an integrated "whole continent" view of North America and improved understanding of the processes that have shaped and continue to shape the continent, and that directly affect our lives.

The elements of USArray described in the following text can, individually and in concert, realize a number of scientific goals and address a range of issues of societal concern. Outstanding Earth science problems that can be addressed with USArray include:

USArray can also serve to advance our understanding of natural hazards throughout North America. USArray will provide spectacular data for understanding earthquake, tsunami and volcano hazards, and will focus academic attention on unresolved problems related to the earthquake cycle, strong ground motion amplification, and the causes and recurrence of volcanism. The flexible array in particular will be a powerful tool for focussed studies of societal hazards such as magma movement around active volcanoes in the Pacific Northwest and movement on faults in seismically active regions. Elements of USArray will also provide information on local and regional scales useful to resource managers to help evaluate groundwater, hydrocarbons, and mineral resources.

2. Rationale for the facility. Consider what our understanding of North American tectonics would be if the only topographic information we had to work with was similar in resolution to modern traveltime tomography images (Figure 1). First-order features such as the division of the continent into a stable continental interior (characterized by fast seismic velocities and low topography) and an active western margin (exhibiting low velocities and high elevation) are resolved, but little more than this gross subdivision is possible (Figure 1a). With increased resolution we begin to see more detailed structure corresponding to the largest-scale tectonic provinces (Figure 1b). Where seismic observations are available, the trends in surface features often correlate with mantle structures. Where we have data, the upper mantle shows considerable variation at lateral scales of <100 km. Areas with little mantle structure correspond to regions with few or no observations. At 1-km resolution in topography (Figure 1c) we see the richly detailed morphology that defines the tectonic subprovinces of North America, with much of this topographic detail representing the manifestation of deeper-seated geologic processes. Clearly, however, our images of deeper lithospheric structure are orders of magnitude coarser than our surface images, and more importantly they lack continuity over the large lateral distances necessary for understanding continental dynamics.

In places, individual experiments and investigations using dense source/receiver geometries have produced spectacular results and new insights of specific, local or regional tectonism (Figure 2) but collectively these results provide only a rough patchwork understanding of the structure and evolution of the continent. It is difficult to tie results from individual experiments, spaced sparsely across the continent, into a coherent picture that ties surface geology, and detailed lithospheric structure to driving forces within the deeper mantle. In numerous experiments we see examples of coupling between the crust and mantle. We see whole crustal faulting at strike-slip plate boundaries and orogen wide crustal detachment faults implying mantle control of the crust in orogenic belts and along plate boundaries. We see crustal recycling and dewatering within subduction zones, and delamination and phase change of mafic lower crust beneath orogenic belts, all providing crustal control on upper mantle structures. The abrupt transition from cratonic to tectonic upper mantle in western North America implies long-lived (1.5 Ga) structural control on modern tectonics (i.e. Proterozoic terrane accretion boundaries have guided Mesozoic-Cenozoic tectonism). The smallest resolvable scale mantle structures in tomographic images of tectonic North America show a strong correlation between surface geology and dynamic upper mantle structure. Moreover the few available higher frequency observations imply that detectable mantle structure exists across the entire spatial bandwidth, the structure of the mantle is as rich as that of the crust.

Nevertheless, we currently lack a coherent understanding of the degree of coupling and mechanisms of stress and mass transfer across fundamental rheological and compositional boundaries in the lithosphere. Our knowledge is even more incomplete when we look deeper at the asthenosphere and lower mantle. USArray will provide a continuous imaging capability that will tie together North America's seemingly disparate tectonic provinces into a coherent model of the origin and evolution of the continental lithosphere.

USArray will also provide data for forefront research on deeper Earth structure from the upper mantle through the core, including the structure of the mantle transition zone, the core-mantle boundary, and the inner core/outer core boundary. In addition to advancing integrated research in basic Earth structure, the facility will improve seismic hazard assessment in the US, and greatly enhance the public visibility of the Earth sciences providing a unique opportunity to impact Earth science education and outreach across the country.

The time is right for an initiative of this magnitude. Significant advances in technology, theory, and imaging capabilities have taken place over the last ten years. The seismological community has gained significant experience collecting large, broad band, high-quality, digital data sets from array deployments of seismometers. Global tomographic images with a resolution of 5° (~550 km) hint at links between shallow and deep Earth processes while advances in reflection profiling allow us to see whole lithospheric structure at scales of less than 100 m. At the same time there is an increased recognition that the next major advances in our field will come from the integration of measurements contributed by a diverse set of Earth science disciplines. Many problems in Earth science, particularly those of the magnitude to be addressed by USArray, are increasing in complexity beyond that which a single discipline can successfully solve. A multidisciplinary approach is required to make links between disparate datasets, to provide constraints for interpreting each dataset, and to develop a complete understanding of the structure of the continent and the dynamics and history of continental development. Finally, advances in information technology now make it possible to manage and manipulate large and disparate data sets, facilitating joint interpretation.

Seismology returns information about the variation in density and seismic velocity of the subsurface. Electromagnetic techniques provide information on electrical conductivity of the subsurface, and geodesy gives the current velocity field of the surface. Petrology and mineral physics provide the data to translate seismic velocity into rock type. Rock mechanics offers the link between strain and the movement of rock. Geochemistry and field mapping provide information on what rock types might be present in the subsurface, and geochronology provides information on the timing and sequence of events in geologic history. Geomorphology and stratigraphy provide classical but still critical constraints on vertical motions, geologic history, and feedbacks between surficial and tectonic processes that shape the surface of the continent. Finally, geodynamic modeling integrates this information in an attempt to understand the plate and local interactions that led to North America’s current structure and predict its behavior in response to the existing forces acting on North America. The combined input from these varied disciplines contributes to a better understanding of the images returned from the USArray geophysical "instrument". Their explicit inclusion in USArray is essential to maximize the scientific return from this project.

The upper mantle plays an extraordinarily important role in Earth dynamics and the three dimensional structure beneath continents has not been resolved sufficiently well to answer fundamental questions. While it is clear that the upper mantle is anisotropic and that this anisotropy probably results from the preferred orientation of minerals controlled by mantle flow, we do not know the details of the anisotropy which can be a powerful probe for illuminating three dimensional mantle flow. Furthermore, the thermodynamics of the two major jumps in seismic velocity at 410 km and 660 km depth, probably resulting from solid-solid phase transitions, are poorly known. The constraints from small, regional studies around the world are confusing and the absolute magnitude of the topography, the lateral scale-lengths of topographic undulations, the sharpness of the boundaries, and how the topography of the discontinuities varies with volumetric changes in seismic velocity are all poorly known. With USArray we have the opportunity to determine the true nature of the mantle transition zone with unprecedented lateral resolution and how this relates to dynamic processes within the mantle, both upwellings and downwellings, and in turn how deeper dynamics is related to surface geology. This would be a spectacular advance for the Earth sciences (see Appendix 2 for details).

The North American continent exhibits a diverse array of tectonic provinces (Figure 3). The western US contains a Mesozoic-Cenozoic mobile belt and orogenic plateau including a Cenozoic collapse structure, a developing rift system, a continental hotspot, and one of Earth's most important ocean-continent strike-slip fault systems. The eastern US contains a Paleozoic continent-continent collision zone, a Mesozoic rift, and a Mesozoic-Cenozoic passive margin. The passive margin developed along the Gulf coast contains one of the best examples of salt tectonics and one the world's largest accumulations of hydrocarbons. In the interior of the continent, Proterozoic collisional belts ring the Proterozoic-Archean craton which is also the site of one of the Earth's largest upper mantle velocity anomalies. The mid-continent also is blanketed by platform sediments, which represent a long and detailed record of what are probably mantle-controlled vertical motions. In the western US, the youngest tectonic activity, operating over the last few million years to tens of millions of years, is responsible for much of the architecture that we observe today. Understanding this recent activity provides insight into plate boundary processes and associated hazards from earthquakes and volcanoes, and also provides insight into processes that have operated over geologic time. However, this recent activity overprints older structures developed during previous plate collisions and rifting events and the existing structures modulate modern processes. Unraveling the complex geologic history of the continent requires understanding both the modern and ancient records. A systematic investigation of the North American continent provides a broad sampling of geologic processes across the widest possible range of scales and ages. In essence, moving spatially across the continent provides a snapshot of continental evolution and continental dynamics through time. As a natural laboratory North America provides a diverse array of processes and structures for investigation, an opportunity for studies with societal relevance (investigation of earthquake and volcanic processes, hazards, and natural resource evaluations), and an existing infrastructure (communication, transportation, and educational) which makes the project feasible.

3. Development of the initiative. Over the past few years, there has been discussion within the seismology community of the opportunities, both scientific and technical, for an initiative to improve the resolution at which we can image the structure of the continental lithosphere and mantle, and to merge seismological results with other Earth science investigations (see Appendix 1). At the same time, there is growing interest in the use of dense deployments of broadband seismometers to investigate deep Earth structure. These discussions culminated in a workshop in March of 1999 in Albuquerque NM. The workshop, jointly sponsored by the National Science Foundation and IRIS (Incorporated Research Institutions for Seismology), was attended by over 90 participants representing a broad spectrum of Earth scientists with representatives from academia, the US Geological Survey, regional networks, and the National Science Foundation.

At this workshop, seismologists and geologists discussed the design and implementation of an ambitious plan to explore, image, and develop an integrated understanding of North American geology and deep Earth structure. Workshop participants helped define the technical components of the USArray facility, identified scientific goals, and discussed an operation and management scheme for the facility. The Albuquerque workshop led to substantial enthusiasm and momentum for USArray and increased recognition that this initiative needs to integrate geological and geophysical investigations into a single unified effort to best achieve its scientific goals.

A second workshop was held in September 1999 in Houston, TX to integrate a diverse group of Earth scientists into the early planning stages of the USArray initiative to enhance its scientific goals, better define its multidisciplinary component, and identify ways in which USArray can best be used to advance Earth science research, education and outreach. The Workshop also included discussions of how the Earth science community can work together with the National Science Foundation to enhance support for our research endeavors - through development of compelling scientific studies, through integrative community projects such as USArray, through clear statements of our long term science goals, and through enhanced public appreciation for Earth science research. This white paper and the attached appendices document the outcome of these two workshops.

4. Outline of the facility. USArray is designed to probe the three-dimensional structure beneath continental North America, using a spatially dense network of high quality seismic stations. It can be thought of as an inverted telescope designed to image with unprecedented clarity the spatial distribution of rock properties, from which we can elucidate Earth structure and test hypotheses about how the lithospheric mosaic that makes up our continent was assembled.

4.1 The geophysical component of USArray consists of three interrelated parts (see Appendix 1 for additional details). The first component of USArray is a transportable telemetered array of 400 broadband seismometers designed to provide real-time data from a grid with dense and uniform station spacing of ~70 km and an aperture of ~1400 km (Figure 4). The array will record local, regional, and teleseismic earthquakes, providing resolution of crustal and upper mantle structure on the order of tens of kilometers and increased resolution of structures in the lower mantle and core-mantle boundary. Approximately 50 magnetotelluric field systems will be embedded within the array to provide constraints on conductivity which translate into temperature and fluid content within the lithosphere (see Appendix 5 for additional details). The transportable array will roll across the country with 18-month deployments at each site (Figure 5). As the array moves, it will systematically and uniformly image structure beneath the continental US. Continuous redeployment will cover the entire conterminous US over a period of 10 years, providing unprecedented 3-D seismic imaging. The result of this experiment will be the first coherent, high-resolution, plate-scale image of the lithosphere and underlying mantle for either the continents or the oceans; it will focus the blurry images presented in figure 1 producing the synoptic image that will form the foundation of a new and comprehensive understanding of continental tectonics. In addition to the investigation of the rich variety of lithospheric structures provided by the complex tectonic fabric of North America, the location of the US with respect to earthquake sources in the SW Pacific will allow USArray to provide an excellent view into the deep mantle. The transportable array, will provide a three-dimensional "volumetric map" of seismic velocities and anisotropy beneath the United States down to at least 1,000 km depth at a uniform 50-200 km resolution. Within this volume of the mantle, the topography and sharpness of the principal jumps in seismic velocity will be mapped out beneath the continent with a lateral resolution of several tens of kilometers. While the initial focus of USArray is coverage within the United States, extensions of the array into neighboring countries and onto the continental margins through collaborations with scientists from Canada, Mexico and the ocean science community are natural additions to the initiative.

Magnetotelluric field systems will be embedded within the array to provide constraints on conductivity which translate into temperature and fluid content within the lithosphere (see Appendix 5 for additional details). The magnetotelluric method is most sensitive to conductors in the crust and upper mantle. Silicate minerals at subsolidus temperatures in the crust are very resistive, so natural electrical currents are attracted to regions of low resistivity. These conductive regions can be caused by solid phases such as graphite and metallic sulfides or oxides, aqueous fluids (especially brines), and partially melt. A common requirement in all of these causes is that the conductive fraction must be interconnected, thus leading also to inferences about fluid migration paths. In the mantle, the apparently simpler mineralogy leads to fewer potential causes of lower resistivities. Resistivities of silicate minerals in the mantle decrease systematically with temperature, allowing estimates of temperature from MT soundings. Seismic properties such as attenuation and compressional and shear wave velocities can be coupled with electrical resistivity to allow discrimination of the type and state of crust. For example, the conditions under which partial melting can occur can be ascertained by the joint use of velocity and resistivity. Interpretation of MT data is truly a 3-D problem but few surveys have the regional distribution of stations to permit development of a regional 3-D model. Coupling MT soundings with the transportable array would provide this needed 3-D regional MT structure and permit a quantum leap in our ability to interpret data. Reinterpretation of existing surveys in this regional context could reveal structures missed in the earlier, 2-D models, and new surveys would be planned with the regional network in mind. An outstanding scientific issue is the controversy over the existence of fluids in the old continental crust. Many MT surveys report the existence of moderately conductive lower continental crust, but petrological considerations appear require a dry lower crust. One criticism of the conclusion that the lower crust is conductive is that MT surveys are most often done to delineate anomalous crust so that perhaps all we can conclude is that anomalous crust is conductive and we do not know the properties of normal crust. An advantage of coupling MT soundings with the "Bigfoot" array would be a systematic sampling of the entire continental US, which is underlain in many areas by old, stable continental crust.

An important second component of USArray is an additional pool of ~2400 instruments (200 broadband, 200 short period, and 2000 high frequency) that can be deployed using flexible source-receiver geometries. These additional portable instruments will allow for high-density, shorter-term observations of key targets within the footprint of the larger transportable array using both natural and active sources. This flexible component of USArray offers exciting opportunities for a variety of focused investigations requiring higher resolution images imbedded within the context of the larger array. Resolution provided by this flexible array will vary from the 60-kilometer scale provided by the transportable array to the tens of meters scale, making it ideal for tying surface geology to deeper structures in the crust and upper mantle.

Many geologic targets are amenable to investigation using the flexible array. Examples include: the structure of seismogenic and creeping faults, delineation of magma chambers beneath active volcanoes, the relationship between crustal tectonic provinces and mantle structure, the shape of terrane boundaries, the deep structure of sedimentary basins and mountain belts, and the structure and magmatic plumbing of continental rifts. In addition, aligning the flexible component of USArray with respect to earthquake sources in the SW Pacific and South America will image the complex and dynamic structure of the core-mantle boundary with unprecedented clarity. These improvements in imaging will provide fundamental constraints on regions of the Earth that contain the boundary layers of mantle convection. The multi-scale observations provided by high-resolution images from the flexible array, deployed in coordination with geological, geochemical, and/or geodetic studies, combined with observations from the larger-scale transportable array will provide an unparalleled opportunity to understand whole systems and can be used to address a wide range of problems in continental geodynamics and tectonics. Examples include imaging and study of the continental arc system in the Cascades from slab to edifice, examination of the deep roots of the North American craton, imaging both ancient and modern orogens and rifts to determine secular variation in continental tectonics, identifying the role of the mantle lithosphere during orogenesis and rifting, and unraveling the relationship between deep mantle convection and surface tectonics.

A third component of USArray is an augmentation of the permanent seismic network in the US operated by the US Geological Survey. Permanent observations from a uniform, high-quality network provide fixed reference points for calibration of the transportable array and a platform for continuous long-term observations. Relatively dense, high-quality observations from a network with uniform spacing of 300-350 km across the continent is important for tomographic imaging of deep-Earth structure and will provide a continent-wide unaliased wavefield for long-period surface wave traveltime and diffraction tomography. Some or all of the stations of the permanent component of USArray will be equipped as expanded geophysical observatories, with GPS receivers to provide direct real-time data on crustal deformation.

 

The permanent component of USArray will be undertaken in coordination with the US Geological Survey complementing the initiative underway at the Survey to install an Advanced National Seismic System (ANSS). The USGS is developing the ANSS to meet its mission in earthquake-hazard analysis and mitigation. To meet these goals requires a densification of the National Seismic System (NSN) with a focus on urban areas with significant seismic hazards. The goals of USArray are to illuminate structure and understand dynamics of lithosphere and deeper mantle, a goal that requires densification of the NSN and uniform coverage of the continent. There is clear synergy between these efforts. The combined efforts and funding of the separate initiatives creates a single integrated network of ~100 high quality stations to meet the goals of both constituencies. All data from this network will be available in a single data stream to both communities. Coordination of these two initiatives is a prime example of interagency cooperation and cost sharing and continues the longstanding relationship and interaction between IRIS/NSF and the USGS on permanent seismic stations.

The GPS component of USArray is an important complement to the seismic instrumentation (see Appendix 4 for details). At least 65% of North America's area has potential "signal" for a continent-wide array of high-precision GPS stations. Even the remaining "stable" area is important for GPS observations, partly because such stations help define a stable North American reference frame for the velocity field, and because it is possible that signals will be detected in the "stable" regions that are beyond our current understanding of displacements and stresses for cratonic regions. A continent-wide array of high-precision GPS stations, with strict configuration control to ensure accuracy, is also an imaging device, but in this case we seek to "image" the continent's present-day surface velocity field. Over the length of the USArray project the velocity field can be processed to give the modern strain field in the plate boundary zone, as well as elsewhere in the continent. With this velocity field, we can constrain neotectonic processes at plate-boundary zones and other regions, contribute to earthquake-hazard estimates, test the limits of plate rigidity, and infer mantle rheology. We can also test hypotheses about how the current lithospheric mosaic responds to plate-scale and local stresses, and investigate enigmatic mid-plate seismicity.

4.2 The Geologic Component of USArray. The various geophysical components assembled into USArray will provide an unprecedented examination of seismicity, seismic-velocity variations, crustal and upper-mantle conductivity, and crustal movements on a continent-wide scale. To translate these images into a better understanding of the structure and evolution of the North American continent will require substantial interplay from various of disciplines within the Earth sciences (see Appendix 3 for details). A strong geologic component to USArray is required to interpret the geophysical images, enable an understanding of continental evolution by providing timing and rate constraints, guide targeting of the flexible component of the array, and provide a multidisciplinary component for studies of problems that can no longer be solved by one technique alone.

The seismic instrumentation employed in USArray will create an image of the current seismic velocity structure of the crust and upper mantle of North America. The seismic velocities of rock however, are affected by their physical state and chemical composition so that temperature, composition, mineralogy, shear fabric, and the presence or absence of pore fluids all play a role in their values. Important ancillary studies must be used to interpret observed seismic-velocity variations. These include: laboratory determination of phase proportions, mineral elastic parameters and rock physical properties at differing pressure temperature, and pore fluid content; correlation of patterns revealed by surface exposures identified during field mapping with patterns observed in seismic data throughout the crust and upper mantle; analysis of xenoliths of deep material brought to the surface by certain types of explosive volcanism to correlate with seismic images and constrain the composition and physical state of the upper mantle; examination of the geochemical properties of magmas generated in the deep crust or upper mantle. Within the last decade, the fields of mineral physics, petrology, and rock mechanics have greatly expanded their ability to measure and understand these parameters, particularly at the temperatures and pressures of the Earth’s interior. However, to realize the full benefit of the seismic images produced by USArray, much work remains to be done in determining high-quality elastic parameters and their pressure and temperature derivatives, in order to translate seismic velocity maps into information about rock-type, physical state, and rheology.

In areas where tectonism and erosion have brought deeper materials to the surface, petrologic and structural analysis provide a wealth of information concerning the history of rock units and the conditions they experienced at depth. Detailed field mapping is a critical constraint for any seismic interpretation of continental structure and evolution. In concert with this, petrologic laboratory experiments that determine metamorphic reaction kinetics are necessary for understanding key components of the orogenic cycle. Geochronologic measurements can provide precise timing for both rock forming and deformational events, and exhumation. Stratigraphic studies, some employing techniques in high-resolution geophysical imaging, can provide a detailed measure of vertical motions, and geomorphic studies can complement these data, particularly in more active regions, and particularly now that methods have been developed to measure rates of erosion using cosmogenic isotopes. Field examination is a critical end-member component of any study of continental structure and evolution, and a program of focused new studies coordinated with USArray deployments will be needed, as well as a systematic effort to compile and make accessible the great deal of data that already exist (see section 4.3).

Xenoliths, random fragments of the deep crust and upper mantle brought to the surface by explosive volcanism, provide critical constraints on what rock types are present in the deeper crust and upper mantle, and permit construction of the compositional stratigraphy and thermal profile of these regions. Complementary but independent constraints on the composition of the deep crust and upper mantle can be obtained through geochemical and isotopic study of magmas, now crystallized and exposed at the surface. Such data can be combined with modern geochronological analyses capable of providing precise ages of formation and time-temperature histories for both xenoliths and exposures of igneous and metamorphic rocks. Such information is critical for understanding the degree to which surface and deep features are connected.

Modern geochronology can address many time scales, from the precise Holocene record now available through dating techniques that employ short-lived radioisotopes to developments that allow uranium-lead ages to achieve precisions of ±1 million year on rocks as old as 4 billion years. At the short end of this time scale, geochronologic techniques can be directed at issues such as the record of movement along faults, the rate of uplift or subsidence in tectonically active settings, and the recurrence interval of active volcanoes. At longer time scales, geochronology can address issues of the rate of continent growth, identify terranes of differing age and history, define the timescale of deformation accompanying continental collision or rifting, and address whether seismic anisotropy in the crust and upper mantle is a response to modern strain fields, such as plate movement, or was frozen in from ancient deformation events.

Geochronology is a technique of similar maturity to seismic imaging. With recent breakthroughs in analytical throughput and cost, and with the precision and range of applications currently available, resolution of most of the important events in the development of the continent is possible, and doing this at a continental scale is imaginable. The quality of continent-scale imaging that will be provided by the USArray instrument can only be fully exploited by a coordinated geochronological effort capable of resolving essentially all the major events involved in the assembly and break-up of North America, and providing rate and timing constraints on both ancient processes and those that are currently responsible for the geologic hazards currently experienced in North America.

4.3 Earth science information system. A shared and critical requirement for all components of the USArray facility is communication (telemetry), data archive, and access and management of diverse data sets that include both geophysical and geological data. A system of data management (archive, distribution, and visualization) that will simultaneously allow integration of geological and geophysical information is critical to scientific analysis. A goal of USArray is to aid in the development of an Earth science information system that can be accessed easily by the broad Earth science community, educators, government agencies, and the interested public. All seismic, MT, and GPS data from USArray will be archived and available in near-real time and integrated with appropriate geologic datasets and provided openly to the research community, educators, government agencies and agencies responsible for hazard assessment.

Development of an Earth-science information system fits within NSF priorities to develop and enhance information technology. Ultimately, such a system could end up as the greatest legacy of the USArray initiative. Consider the state of the Internet, just some 10 years ago (or less, from many people's perspective). The Net was a rich source of information, but it was also a relatively hard-to-use, hard-to-navigate, hard-to-integrate jumble of ftp, telnet, gopher, wais and other sites and protocols. The arrival of the world-wide web in a short time transformed and democratized the Internet, and initiated a revolution in information exchange. In the natural sciences today, although we have many data and many tools for manipulating and visualizing these data, we have few if any software tools analogous to a web browser that can seamlessly bring environmental data to our attention and use. This system would be highly transportable to other fields of science and management.

The IRIS Data Management Center (DMC) has a well-established system for distribution and archive of large volumes of digital waveform data over the Internet. IRIS has also agreed to archive and distribute magnetotelluric data from EMSOC (the University Consortia for ElectroMagnetic Studies Of the Continent). Both IRIS and UNAVCO can archive and distribute GPS data. The USGS is developing archives of potential field data, geochemical data, digital elevation, remote sensed data, and map data that are accessible via the Internet, but these efforts are in various states of completion. A concerted effort will be required to bring these data bases to a level where they can be used routinely in Earth science investigations. An attempt also needs to be made to develop a data base of drill hole and industry reflection profile data. A growing realization in information systems is the need to provide end users not only with access to data, but also with software to aid data integration and manipulation. Software for data exchange, integration, and visualization tools need to be developed. The huge data stream to be produced by the seismic component of USArray requires establishment of a dedicated data management facility. With today's technology this "facility" could be distributed, but it needs to have underlying, standards and coherence to accomplish the research and applied goals of USArray. Such a facility would vastly improve the ability to combine geologic and geophysical information to meet the scientific goals of the USArray project as well as serve as a tremendous resource to the whole Earth science community.

A considerable amount of information already exists from decades of geological, geochemical and geochronological studies of North America. Much of these data, however, are in a form that is not readily compared with other information, such as the geophysical images to be returned by USArray. A large portion of this geological data exists only in print-published tables and maps, or in electronic format in an individual scientist’s computer. For several large categories of Earth science data, no data exchange standards exist to facilitate information sharing. To maximize the availability of the existing information and its contribution to the USArray effort, there is a need for the assembly of these disparate data into a consistent format that can be accessed and used by the whole geoscience community for multidisciplinary studies. The need for format standardization and for data archival is obvious. One need only consider some potential applications to understand the utility of such an easily accessible database. For example, one might wish to overlay a map of seismic velocity in the upper mantle with a compositional map of basaltic volcanism in the western US. Similar efforts in the ocean sciences revealed a global correlation between degree of partial melting and elevation of ocean ridges that was quickly translated into dynamic models of the temperature structure beneath the ridge and its relation to spreading rate and volcanic productivity.

Full interpretation of the information returned from the USArray geophysical instrument will be aided greatly by ready availability of geological data for North America. Given the data volume that will be archived for the USArray seismic data alone, the addition of a geological data archive complementing that under development by the USGS will be only a small additional data volume that will have huge potential benefit, once data exchange formats and protocols are established. For USArray, an important part of the data management mission is to facilitate the adoption of standards for data exchange by the geologic community and to facilitate the transcription of existing data into these standards.

4.4 Education and Outreach. A highly visible, science-driven initiative such as USArray provides a spectacular opportunity for a comprehensive education and outreach program (see Appendix 4 for more details) that will extend across the country and continue throughout and beyond the lifetime of USArray. USArray capitalizes on the public's natural interest and excitement in earthquakes and Earth science and can make these subjects relevant on a region by region basis as the array is deployed across the country. The USArray initiative offers many possibilities for outreach and education to reach the general public, K-12 students and teachers, and Earth science students and professionals at all levels. We anticipate a comprehensive program that tries to convey both the new scientific results that emerge from the USArray national scientific effort, and perhaps as importantly, the nature of our scientific method. This effort will be strengthened as USArray moves from region to region highlighting both important regional questions and an emerging (and changing) continental scale picture.

USArray will play an important role in educating the public in the Earth sciences, and science in general. As USArray moves across the country it will be coordinated with a comprehensive educational and outreach program highlighting both overarching and regional Earth science issues (hazards, structures, resources) and links between Earth science and society. Specific initiatives will include development of Virtual Seismic Network (VSN; see Appendix 1.3), software for use by educators in K-16 educational programs, museum displays, public and commercial media programs, teacher workshops, and educational materials and activities linked to array data and Earth models available over the Internet. As a result of the National Science Education Standards and state standards, many K-12 teachers are having to teach Earth science for the first time. USArray can help get solid content into the curriculum through a real experiment. Students can learn science by doing science. Programs will be designed and targeted to engage communities in the USArray endeavor before, during, and after passage of the array through specific regions of the country. The USArray initiative can reach national audiences through PBS/Discovery Channel programs, an "Earth Minute" on NPR, and regular feeds to Weather Channel. The latter could include earthquake reports, real-time data feeds, and an "Earthquake of the Day" and could highlight geologic structures and hazards that are unique to the region currently occupied by the array.

4.5 Operational Mode, Community Input, Science Support, Partnerships. USArray is a science driven initiative. The facilities developed through this initiative are community resources. To maximize the scientific return from USArray and maintain the flexibility to fund the best science, the operation, management, and use of the facility must incorporate: high-quality, peer-reviewed, PI driven research; quality controlled data acquisition and archives; standardized data exchange formats; and open and timely access to data. These requirements must be met with an attitude of cooperation within the Earth science community using the data and a commitment to service from the USArray facility operators.

The flexible component of USArray will provide high spatial resolution images of features in the lithosphere and deep mantle. Unlike the transportable and permanent components of USArray, the flexible array will not be capable of imaging all of North America. Consequently, deployments of the flexible array will be selected through proposal competition with the expectation that specific geological problems requiring multidisciplinary investigations encompassing a wide spectrum of the Earth sciences will be a critical component of a successful proposal. Many geologic and deep Earth targets are amenable to investigation with the flexible array. To choose the best targets from those proposed for the flexible array will require balanced review from a wide segment of the Earth science community.

In order to engage the broadest possible participation of the Earth science community and to maximize scientific return as USArray sweeps across the country, regional workshops need to be held to define regional science objectives and to engage a broad spectrum of scientists in the initiative. Three workshops, one each focusing on the western US, central US, and eastern US should be held within the first year of the project. The purpose of these workshops is to inform and educate the community about the capabilities of USArray, to begin to engage and enfranchise a broad spectrum of scientist and educators, to begin to facilitate and encourage cross disciplinary investigations, and to begin to develop the infrastructure, contacts, and databases to maximize the use of USArray.

Additional regional workshops will be held 2 Years in advance of the transportable array deployment in an area. The purpose of the second set of workshops is to focus on important targets and investigations within the array footprint, to foster early multidisciplinary exchange, and to identify and utilize existing infrastructure, contacts, and databases in each area. If done successfully this will ensure the development of compelling, high-quality science proposals to make use of the USArray facility. Additional workshops will take place 6 months in advance of deployments after proposals are funded for project coordination, and 1 year after deployment for science integration.

The Earth science community, working in concert with NSF EAR and other appropriate agencies, will need to find appropriate levels of research support for the broadly based science envisioned for USArray facilities. This scientific research will include: 1) analysis of data from the permanent and transportable array deployment, 2) use of the flexible instrumentation, often for multidisciplinary projects, but potentially for single investigator experiments as well, and 3) geologic studies that are integrated with the USArray deployments.

To truly succeed as an integrated Earth science project it will be necessary for national and state funding agencies to provide a higher level of research support than is currently available. Possible models of support to explore include those employed by the Ocean Drilling Program, the NSF Science and Technology Centers, and focused NSF programs similar to Margins and Ridge. Regardless of the final mechanism (or set of mechanisms) used to support science, it is essential that an appropriate peer-review system be established within NSF to evaluate proposals submitted to use the USArray facilities. The goals are 1) to maintain the delicate balance between excellent single-investigator science and larger multidisciplinary projects that have made the US universities the best in the world, 2) over the next ten years, to forge a truly national initiative to image North America and 3) to facilitate a gradual change to more integrated science that will last well beyond the time span of USArray. This will require a pool of reviewers familiar with the objectives of the initiatives as well as the fundamental scientific issues. This will also require an NSF panel structured to accommodate the potential multidisciplinary nature of the investigations and to balance discipline-specific research and multidisciplinary research. The review process needs to include mechanisms to facilitate multidisciplinary investigations, provide oversight, long-term perspective, and fair review, all the while encouraging the individual creativity, insight, and innovation that comes with PI-driven research. Finally, while we seek to maximize the potential scientific return from coordinated studies of the North American lithosphere and deeper Earth, we need to continue to fund the best science wherever these investigations take place around the globe. An appropriate balance and level of funding must be found and maintained.

An initiative of this scope obviously requires a number of partnerships between the academic Earth science community and a variety of organizations including: the National Science Foundation, US Geological Survey, regional seismic networks, State Geological Surveys, and academic consortia (IRIS, UNAVCO, and EMSOC). International partnerships and collaborations with industry will also be important as the project matures. These partnerships should be explored and developed at the earliest stages possible.

4.6 Beyond Research. The resources and results from USArray will find wide application in a variety of issues of growing societal need by advancing our understanding natural hazards and natural resources throughout North America. The opportunity for these applications should be included in the early planning and implementation stages of USArray. The flexible array in particular will be a powerful tool for focussed studies of societal hazards such as magma movement around active volcanoes in the Pacific Northwest. The flexible array will also extend the instrumentation available for attacking a wide range of problems in fundamental studies in earthquake dynamics, fault zone imaging, and characterization of fault zone properties. The instruments of the flexible component of USArray will also provide a unique resource for basin studies, in the tradition of those carried out recently in Seattle and Los Angeles important for characterizing the potential for strong ground motions in urban areas and movement on faults in seismically active regions. Many such studies are already underway as part of efforts by the USGS and other organizations, but partnerships between these studies and USArray will be especially powerful because USArray offers additional instruments for process-oriented studies and the ability to link small scale tectonic activity through to large scale driving forces.

Different elements of USArray will also provide information on local and regional scales useful to resource managers. The high frequency instruments in the flexible array will be available for three dimensional regional groundwater resource assessment and subterranean waste treatment and management studies. For example, studies of the detailed geometry of aquifers around major western metropolitan areas are underway, but will be strengthened by the ability to study specific basin bounding faults, ground subsidence, and basin velocity and density structure. Exploration for oil and gas and mineral resources has traditionally been the job of private industry, but using the Canadian Lithoprobe Experiment as an analog, the flexible array can used to help improve our understanding of subsurface structures beneath basins. The power of USArray in the realm of resource issues will be fruitful partnerships between private industry and principal investigators. The flexible and transportable arrays together will provide a detailed tectonic framework of the continent that will be an invaluable resource for mineral and energy exploration.

As USArray systematically traverses the continent, there will be natural opportunities for a variety of interactions with State agencies involved in research, education, public policy and resource assessment. Much of the data and results from the USArray program will be of interest to state geologists in hazard evaluation (seismic, volcanic, landslides), and assessment of mineral and water resources. Crustal and lithospheric imaging with USArray will add to our fundamental knowledge about Earth structure and will feed directly to state geological mapping projects.

5. Summary. A national program on the scale of USArray can catalyze solid-Earth science research and help organize this discipline's contribution to Earth-system science in North America. The scientific and organizational structure underlying this multidisciplinary facility can serve as an umbrella for Earth science studies in continental structure, dynamics, and evolution, and studies of the deeper Earth. It can:

Finally, perhaps the most exciting aspect of this initiative is the prospect of unanticipated discovery, the unveiling of results and insights we can not yet imagine. We need only look back 10 to 15 years to see the dramatic evolution of our discipline by comparing our views of Earth structure and dynamics then and now (Figure 6). Advances in facilities, techniques, theory, and data integration have made and will continue to make this dramatic progress possible. USArray presents an exciting opportunity for growth and development of new ideas. There is no doubt that a facility like USArray and our approach to solving scientific problems will evolve over the 10-15 year period of its operation. Novel ideas will emerge and new research targets will be identified which will require new theory, analysis techniques, and research tools. Our understanding of Earth structure and dynamics will undoubtedly be more comprehensive and complete and perhaps as different from our view today as that of 10-15 years ago.

Recommendations. The following actions are required to achieve the science goals and implement the USArray concept as outlined in the previous pages. Where appropriate cost estimates have been included in the table below. Subsequent to the first USArray workshop, Program Officers of the Division of Earth Sciences (EAR) united several pending facilities initiatives into a single integrated effort known as "EarthScope - A Look into Our Continent". EarthScope includes USArray and initiatives for a Plate Boundary Observatory, the San Andreas Fault Observatory at Depth (SAFOD), and Interferometric Synthetic Aperture Radar (InSAR). Collectively these facilities will provide an unprecedented set of observational capabilities for the Earth sciences. Many of the recommendations listed below crosscut the individual facilities and should be considered in the context of an integrated EarthScope Initiative.

 

 

USArray Initiative: Cost Estimates

Facilities

Core MRE Request

One Time Costs

 $58.12 million
Seismic Observatory Equipment acquisition and initial installation: transportable array - 400 bb stations; flexible array - 200 bb stations, 200 sp stations, 2000 high frequency stations; permanent network: augmentation of USNSN coordinated with USGS to bring total network to ~100 stations $56.8 million
MT Instrumentation Equipment acquisition and initial installation: 50 MT instruments $1 million
GPS Instrumentation Equipment acquisition and permanent installation: 16 geodetic quality receivers installed with permanent network $ 320K
Additional Facility Needs

One Time Costs

 $2.1-10.1 million
Geology Infrastructure 4 discipline specific workshops @ $25K/workshop to evaluate discipline specific infrastructure needs $100K
Earth Science Information System 5 discipline specific workshops @ $25K/workshop; $275K initial evaluation study $400K
Mass Store for waveform archive $1.6 million hardware(mass store) $1.6 million
Additional Geodetic Observations coordinated with transportable seismic array 400 GPS receivers and antennas integrated with transportable seismic array, monuments for 1000 sites across the continental US $8 million
On-Going Facility Support

Estimate

 $9.1 - $10.7 million
Seismic Observatory Includes: installation, operation, maintenance, data archives $7-$8 million per year (estimate)
MT Instrumentation Includes: installation, operation, maintenance, data archives $200K per year (estimate)
Geology Infrastructure additional technical support for existing labs @ $60K/tech/yr for 20 labs $1.2 million per year (estimate)
Earth Science Information System Technical support, development of tools and visualization software, data storage devices $500K - $1 million per year (estimate)
Operation of additional GPS instruments as part of transportable array Includes: installation, operation, maintenance, data archives $200-$300 K per year (estimate)

Educational Opportunities

Education and Outreach See appendix 4 for details $1 million per year (estimate)
     
     

Science Support

 

Potential.

 $6-$18 million
Science funding for both discipline specific and multidisciplinary science 10-20 single investigator proposals @$100-$150K/yr each per investigator. 10-15 multidisciplinary proposals (@$500K-$1M/yr/proposal) $6-$18 million per year (estimate)*

*Will be driven by proposal pressure and balance between NSF EAR programs and initiative. Will likely evolve through time.

 

 

 

 

Figure 1. Topography/tomography at different scales. (details in figure caption: the linear Sierra Nevada/Great Valley structures are colinear with mantle anomalies, the mantle beneath the Snake River Plain and Yellowstone hotspot is slow, Colorado Plateau has relatively fast mantle ringed by low velocities in the basin and range and Rio Grande rift system, individual basin and range structures marking extension, volcanoes in the Pacific northwest, a crustal welt marking the Mendocino Triple Junction, the track of Yellowstone hotspot, and individual fault traces).

Figure 2. Examples: Snake River Plane results (Deuker/Sheehan/Humphreys, Shear Wave splitting results (Fischer), offsets in Moho at SAF (Levander, Liu and Helmberger), example of reflection imaging and seismicity (Shaw and Shearer), others…

Figure 3. Tectonic provinces with array footprint

Figure 4. Array and example of multiple scale of observations at Parkfield, integration of regional networks (good figure caption will be required).

Figure 5. Snapshots of deployments, plotted above tectonic provinces or topography.

Figure 6. Earth structure (whole earth, subduction zone, and rift) from old intro text book compared to current intro text book.

 

Appendices. The following appendices provide additional details and science justification beyond that which is included in the main body of the white paper for the various geophysical and geologic components of USArray (particularly Appendices 1-5). In some cases, they outline extensions of USArray that may be cost effective for achieving additional science goals (Appendix 3: heat flow measurements, Appendix 4: adding a GPS component to the transportable array). Appendix 6 describes an innovative and comprehensive program for an Education and Outreach program to accompany USArray. Appendix 7 describes OBS facilities that could be used to extend USArray observations onto the continental margins. Finally, Appendix 8 includes overhead slides prepared for presentations on the USArray initiative.

Appendix 1: The Seismic Component of USArray

1.1 History of the USArray Initiative

1.2 USArray Imaging Science Technical Specification

1.2.1 Earthquake Detection and Frequency

1.2.2 Resolution Measures for Direct Wavefield Imaging

1.2.3 Direct Teleseismic Imaging with USArray

1.2.4 Teleseismic Imaging using Surface Waves

1.2.5 Active Source and Microearthquake Seismology

1.3 Development of a Virtual Seismic Network

1.4 Station Operations within USArray

Appendix 2: USArray and the Deep Earth

Appendix 3: The Geologic Component of USArray

Appendix 4: The GPS Component of USArray

Appendix 5: The Magnetotelluric Component of USArray

Appendix 6: Education and Outreach

Appendix 7: USArray and Continental Margins

Appendix 8: Synergy with other EarthScope Initiatives

Appendix 9: Viewgraphs/Overheads

 

Appendix 1: The Seismic Component of USArray

1.1 History of the USArray Initiative. Over the past few years, there has been discussion within the seismology community of the opportunities, both scientific and technical, for an initiative to improve the resolution at which we can image the structure of the continental lithosphere and mantle, and to merge our seismological results with other Earth science investigations (see table below). These discussions have largely been driven by the desire to address outstanding questions in continental dynamics, evolution, and structure, questions detailed in the CD 2020 report and again at the ILIAD Workshop in Taos NM in 1994. The concept of a continental scale seismic observatory was independently suggested at the IRIS Workshop in Glendale CA and at the ILIAD Workshop in Taos in 1994. It received further discussion at more recent IRIS Workshops, including the IRIS Instrumentation Workshop in Santa Fe in 1997, and the annual workshop in Santa Cruz in 1998. There is also growing interest in the use of dense deployments of broadband seismometers to investigate deep Earth structure. USArray is a proposed facility for investigating the structure and evolution of the North American continent and providing enhanced resolution of the Earth's deep interior. USArray is a logical step in a sequence of recent successes imaging the internal structure of the Earth using IRIS PASSCAL and GSN instrumental facilities.

 

These discussions culminated in two workshop in 1999, one held in March in Albuquerque NM, and a second held in September in Houston TX. At the first workshop participants helped define the technical components of the USArray facility, identified scientific goals, and discussed an operation and management scheme for the facility. The Albuquerque workshop led to substantial enthusiasm and momentum for USArray and increased recognition that this initiative needs to integrate geological and geophysical investigations into a single unified effort to best achieve its scientific goals. The second workshop continued the development of scientific goals, better defined the multidisciplinary component of USArray, and identified ways in which USArray can best be used to advance Earth science research, education and outreach.

 

At the same time these discussion are proceeding, there is increasing encouragement for the research community to work with the National Science Foundation to identify exciting directions for fundamental research and participate in the development of new programs. There is also increasing emphasis, from all federal research agencies, for programs that have an integrative and multidisciplinary approach, and which acknowledge the transition from basic research to the support of societal needs.

 

Much of the research in Earth Sciences depends heavily on observational systems. Through the Major Research Equipment (MRE) account, NSF has established a Foundation-wide program for the acquisition of expensive facilities that would be beyond the normal support of individual Directorates. The NSF Earth Science Division has encouraged IRIS, and other facility programs, to explore whether major advances in Earth Science research could emerge from significant improvements in our observational capabilities.

 

The USArray concept has thus developed as:

 

Brief History of Development of USArray Concept

  • 1993-94 - In discussions leading up to the IRIS-2000 proposal, Guust Nolet suggests the idea of a "Lehman telescope" - 1000 instruments over the US at IRIS ExCom meeting.
  • April 1994 - Glendale IRIS Workshop - further discussion of a continental scale array.
    		•
  • Nov 1994 - ILIAD Workshop - Art Lerner-Lam and others detail MATILDA (waltzing array of broadband seismometers across the US).
    		•
  • Oct 1997 - IRIS 10th Anniversary Celebration - Peter Shearer presents map of 1000 seismograph stations across the US .
    		•
  • Nov 1997 - IRIS Instrumentation Workshop - Santa Fe - Earth Cube poster (Levander, Humphreys, Park, Zandt, Shearer).
    		•
  • Dec 1997 - AGU - Dan Weill encourages people to think about major initiatives.
    		•
  • January 1998 - IRIS ExCom meeting - USArray identified as a "companion to PBO", IRIS PASSCAL NSF MRI proposal for proto-type discussed.
    		•
  • March 1998 - IRIS GSN committee discusses contributions to Array USA - produces white paper.
    		•
  • May 1998 - IRIS ExCom letter to NSF EAR Director Ian Macgregor describes PBO and ArrayUSA initiatives - Earth Cube poster included.
    		•
  • July 1998 - Santa Cruz IRIS Workshop - USArray (Ekstrom and Humphreys) and PBO (Silver) initiatives presented and discussed.
    		•
  • Oct 1998 -EAR - Dan Weill- starts moving MRE idea for USArray forward.
    		•
  • January 1999 - USArray (Ekstrom et al., IRIS Newsletter v16, n2, 2-6) and PBO (Silver et al., IRIS Newsletter v. 16, n2, 3, 7-9) articles in IRIS newsletter.
    		•
  • March 1999 - USArray Workshop, Albuquerque, NM (90 attendees), USArray Steering Committee formed.
    		•
  • April 1999 - EarthScope Initiative formed and MRE proposal developed by NSF EAR.
    		•
  • June 1999 - USArray EOS article (Levander et al., EOS, 80, 245, 250-251).
    		•
  • September 1999 - USArray Workshop II, Houston TX., 60 attendees.
    		•
  • November 1999 - USArray GSA Today Article (USArray Steering Committee, GSA Today, v9, n11, )
    		•

    1.2 USArray Imaging Science Technical Specifications

    Prepared by the USArray Technical Specifications Subcommittee: Alan Levander, Gabi Laske, Guy Masters, Frank Vernon, Peter Shearer, Rob van der Hilst.

    As currently conceived the USArray will consist of three components: (1) An augmentation of the US National Seismic Network carried out in coordination with the US Geological Survey to provide a reference array spanning the contiguous US and Alaska, (2) a transportable array of 400 portable, 3-component, broadband seismographs deployed on a uniform grid, informally called "Bigfoot", and (3) an additional ~400 portable, 3-component, short-period and broadband seismographs and ~2000 single channel high frequency recorders for active source and earthquake aftershock studies which are termed the flexible array. Seismic data from the USArray, when fully operational, are expected to exceed 5 Tbytes/year, roughly an order of magnitude greater than the current GSN.

    Below we attempt to begin quantifying the imaging capabilities of USArray with the various common seismological analysis tools, and with different configurations of the flexible component of USArray. We consider the seismic array response characteristics, and resolution obtainable with the Bigfoot array alone, and the Bigfoot array when deployed with the flexible array. The permanent component provides a common base level for referencing measurements made with the other two components, essentially knitting together the other observations. It also is the only part of USArray that can provide an unaliased wavefield for long period surface waves (T> 100 s) across the U.S. This period band is below that which the portable arrays are likely to sample well.

    The transportable (Bigfoot) array design calls for 10 years of continuous deployment at ~2000 sites. Using this criterion, an instrument interval of ~70km, and site occupation of 18 months allows for a continuous deployment which will cover the ~8 x 106 km of the lower 48 states in 10 years, including the 3 year startup as USArray is brought online. The Bigfoot array will occupy ~2 x 106 km2 with a 70 km station spacing. We note that as the operational plan for the transportable array requires that it roll continuously, the footprint changes almost daily. Nonetheless, it is useful to consider the problem as one of a static footprint. Flexible component experiments are envisioned to study key tectonic targets within the very large footprint of the transportable array using the most appropriate instrument configuration to address each target.

    Key concerns for the seismological community are the broadly defined array response characteristics of USArray. Magnitude detection threshold of earthquakes and earthquake occurrence statistics can be used to estimate the number of useable signals that can be recorded within each transportable array footprint. Similarly, resolution of structural details in the Earth using direct wavefield back propagation and imaging, tomography, and array-forming capabilities for the transportable array are easily calculated. When combined with the instrumentation provided by the flexible array the structural resolution and event identification of the combined array can increase dramatically depending upon the deployment strategy, but making quantitative statements regarding resolution and event detection becomes more difficult. In the following we attempt to use general guidelines.

    1.2.1Earthquake Detection and Frequency

    Resolution within any Bigfoot-print will depend not only on array response characteristics, but also on global and regional seismicity as USArray rolls across the country. The earthquake magnitude detection threshold is also spatially variable and is determined from empirical relations and comparisons to existing arrays. Based on previous broadband experiments across the western US, we observe that a reasonable minimum detection threshold for teleseismic events is about mb = 4.5. Considering the proposed instrument deployment time of 18 months per site, there will be good

    coverage from all azimuths (see figure 1) for tomographic imaging. A more restrictive requirement will be for receiver function studies and shear wave splitting studies which will require mb >= 5.5 from deeper earthquakes (figure 2) which will have order 50 events. Other studies will use surface waves to image the continent. These events

    are larger and we can expect several hundred events recorded at each station with reasonably good azimuthal coverage (figure 3). Finally USArray can expect to record several thousand regional events as well as local events for deployments in the western US (figure 4).

    The 1996-1997 PASSCAL broad-band array deployment in Lodore, Wyoming, provides one useful guide to the numbers of events detectable within a given transportable array deployment. During the 11 month Lodore experiment, the PASSCAL array recorded more than 2,100 earthquakes and quarry blasts for which travel times were picked, of which 920 events were regional (between 1 o and 30o) and 475 earthquakes were of mb >= 4.5 or larger at all distances. The 33 station array yielded 78,000 travel times. Scaling these numbers for 18 months gives about 3500 events of all types, 1500 regional, and ~775 mb >= 4.5 earthquakes. Assuming a total of 2000 sites deployed for 18 months each yields ~7,500,000 travel times for the transportable array, and double that number including the continuous deployment of 400 instruments from the flexible array.

    Of primary importance for structural imaging of the mantle is the number of teleseismic events recorded within each USArray footprint in the distance range 30o to 100o, the distance range for upper mantle P- and S-wave receiver function analysis, and an important range for mantle tomography (Figure 5a). Figure 5b shows the number of mb=5.5 and greater events for teleseismic P-waves at these distances for a reference station in the middle of the country (260oE, 40oN), using the PDE catalog and 1000 random start times between 1976 and 1986. (The PDE catalog is incomplete below mb=5.5, but USArray will record significant numbers of smaller events as well). The shaded region shows the 5% and 95% levels, the solid line shows the median number of events. So, for example, to have 95% confidence of recording 100 mb>=5.5 events, a deployment period of about 14 months is required. The planned 18 month deployment period at each site will provide somewhere between 150 and 260 earthquakes with mb>=5.5, and over 1500 earthquakes with mb>=4.5. This is roughly comparable to the experience with the Lodore array, where the number quoted only included events for which travel time picks were made. Adopting a conservative detection value of m>= 5.0 value, and using a median value of 600 earthquakes implies that the ~2000 sites occupied by the transportable array will produce ~2,400,000 P and S travel times for mantle tomography studies. If the 400 element flexible component array is continuously deployed this number doubles.

    In comparison to the current ISC database for P-wave arrival time picks from all events of all sizes on the globe recorded at US stations in the past 35 years, USArray can expect to produce about 15 times as many P-wave travel times (7 times for the Bigfoot array alone) as is currently incorporated in the entire data set for the past 35 years (Engdahl, Van der Hilst, and Buland, 1998). For events of mb >= 5.5, USArray will produce 3 times as many travel times as the record for the past 35 years. More importantly the distribution of stations for USArray will be much more complete and USArray will produce almost 150 times more S-wave picks than are currently available, therefore for joint interpretation of P and S times we will again have almost 150 times more data.

    These data will be of significantly higher quality than the existing dataset, due to the ability to apply signal processing techniques to measure signal coherence across the seismic arrays, reducing error in travel-time picks. The consequences for mantle tomography are a greatly improved resolution of bulk elastic properties, particularly when coupled with the large number of regional and local earthquakes expected from the western margin of North America (Figure 4). Regional earthquakes provide many more nearly horizontal travel paths, improving vertical resolution in the upper mantle, a continuing problem in teleseismic studies. Local earthquakes in western North America can be used to broadly infer the local crustal structure and can provide a deterministic basis for local station corrections for the teleseismic signals from the transportable array. This will be particularly true when tied to detailed studies utilizing the transportable array. The 400-element array with 70km station spacing can be used as a phased array by removing slowness-dependent static corrections for each station, providing high resolution imaging of the deep interior of Earth, as discussed in another Appendix 2.

    The spatially uniform array will permit detailed Q (anelastic attenuation) studies and shear wave splitting studies in the upper mantle given the very high density of consistent amplitude information USArray will provide. A unified tomographic analysis of very dense amplitude and traveltime data will provide a new level of information on P and S velocity, velocity anisotropy, and Qp, and Qs, permitting a high degree of confidence in the translation of seismic properties to lithology, physical state, and composition.

    Receiver function analysis is best done with deeper earthquakes, minimizing waveform complications arising from near source structure. Figure 5c shows the number of M >5.5 earthquakes from greater than 100km for the same reference station at Figure 5a. For the 18 month USArray deployment each site will also record at least 30 to 60 earthquakes suitable for receiver function analysis (note that this is a yet more conservative value than that for traveltimes). Using the median value of 38 earthquakes, and assuming two passbands for receiver function analysis implies a possible dataset of ~304,000 receiver functions across the entire US from the combined transportable and flexible arrays. If the useable data range extends to magnitude 5 earthquakes, this number increases to 1,200,000 receiver functions. This data volume is approximately 4 times that of the data volume used in image formation from recent continental margin reflection surveys extending 100 km with a 12.5 meter horizontal resolution. It is ~150 times the data in the final processed reflection image from such a survey. The size of this data volume underscores the need for more elaborate data handling procedures than are now available to the average US seismologist engaged in receiver function analysis.

    1.2.2 Resolution Measures for Direct Wavefield Imaging

    The resolution of structural details in the crust and mantle beneath North America is quantified by calculating the point spread functions for body and surface wave travel-time tomography, and for direct wavefield imaging with teleseisms and receiver functions. Direct wavefield imaging is dependent on the temporal bandwidth of the source probe, and the spatial bandwidth of the observations. The former is determined by our source type (teleseismic signals, microearthquakes, vibrating trucks, or explosives), and the latter by the geometry of the recording system, particularly the station interval and the total array aperture.

    Traditional resolution estimates for back propagated wavefields for discrete targets are half of the wavelength of the center frequency of the probe, a result from classical scattering theory, modern imaging theory based on the assumption of a perfectly known background velocity field, and a spatially unaliased wavefield. Using similar assumptions, the resolution of variations in depth to an interface are between one quarter and one eighth a wavelength. For receiver functions we assume P->S conversion at an interface or from a scattering point at some depth z, which appears as an S transmission or diffraction at the Earth’s surface.

    In order to back propagate any wavefield , the spatial velocity field must be known with a fairly high degree of accuracy, some estimates are as low as O(3l0) for petroleum exploration in complex structure (Versteeg, 1992), where l0 is the center wavelength and v(x,y,z)=f0l0(x,y,z). The sensitivity of the receiver function imaging on the degree of specification of the velocity is currently not established.

    The wavefield must also be spatially unaliased and completely recorded across at least the first fresnel zone along the Earth’surface for a diffraction point at the target zone projected to the Earth’s surface along the imaging rays. The spatial aliasing requirement is

    At the Earth’s surface, where lmin is the minimum wavelength in the bandwidth (v=fminlmin), qfresnel is the incidence angle at the first fresnel radius at the Earth’s surface, and Dx is the station interval in USArray. As a first approximation we can use the spherical Earth models, in which the fresnel requirement is a function of depth and the velocity variation with depth. The unfortunate restriction that Dx = lmin (z=0)/(2sin qfresnel) is somewhat balanced by the relation that

    And v(z) is an increasing function of z, i.e., ray bending provides a greater range of illumination angles than straight rays. Equivalently we can say that the angular bandwidth for image reconstruction at the target is greater than the angular bandwidth at the surface. The simplest measure of the Fresnel zone, f r, is given using half wavelength interference and assuming a constant velocity and straight rays above the diffraction point The constant velocity can be an rms or average velocity. This gives a rule of thumb for the fresnel radius at the Earth’s surface as a function of wavelength at the target:

    2fr is the recording aperture required for image reconstruction, from which we can determine the maximum depth for which different components of USArray will provide an adequate aperture.

    1.2.3 Direct Teleseismic Imaging with USArray

    Generally two period bands are used for receiver function analysis, the 0.5-4s, and 10-30s band. If we assume that we have continuous spatial sampling then for teleseismic signals in the 0.5-4s band, which have proven to be useful for imaging crustal and upper mantle structure, a conservative measure of maximum theoretical resolution is about 2.5-5 km for discrete objects and 1-2.5 km for interfaces in the crust and upper mantle. Similarly, the 10-30s period band has a maximum theoretical resolution of ~45 km for discrete objects in the upper mantle, and about 10-20km for interfaces.

    Since USArray does not provide continuous spatial sampling the array response characteristics must be considered, which are determined by element spacing, total aperture, and the geometry of the elements. To consider aperture we calculate the fresnel zone for curved rays in a radial Earth model using the range and travel time integrals:

     

    Where p is ray parameter and b is shear velocity as a function of radius in the Earth, r, and ztarget is the depth of the diffraction/conversion point. Again using a half wavelength criterion the Fresnel radius vs period is plotted in Figure 6a for targets from crustal to transition zone depths. The velocity model is a hybrid of PREM for the mantle (Dziewonski and Anderson, 19XX), and the Christensen and Mooney (1995) average continental crustal model. Figure 6b is the point S-wave diffraction response for USArray in one of the short directions (EW, NS), using an exploding diffractors model, similar to that used in reflection seismology to estimate aperture requirements and resolution obtainable with migration for a given experiment geometry. Diffraction points are at 800, 670, 410, 300, 200, 100, 50 and 25 km, and the bandwidth is from ~ 7.5s < T < 30s.

    Figure 6 shows that each footprint of the transportable array, with an aperture of ~1400km in the short direction, is adequate for imaging targets to the bottom of the transition zone, and has adequate spatial sampling for upper mantle imaging to depths of ~100-200km in the 10-30s period band, but is inadequate for direct wavefield imaging in the uppermost mantle and crust. At shallower depths, direct wavefield imaging will require the denser spatial sampling provided by the flexible component array. An interval of 10 km for flexible array experiments shows that the first fresnel zone will be just adequately sampled for targets deeper than 25km, and well sampled for targets below 50km in the period range 0.5-4.0s.

    1.2.4 Teleseismic Imaging using Surface Waves

    An unprecedented amount of high-quality surface wave data will be obtained from USArray. A typical 18 month period provides about 145 suitable events for recording intermediate-period surface waves (120-20s) in the continental US. Event selection criteria include the epicentral distance range (< 95 deg), the source depth (< 200km) and surface wave magnitude (Ms> 5.3). Figure 7 shows a typical event distribution for the period of July 95 through December 96 (taken from the Harvard CMT catalog). Though the number of events in a given 18 month period has fluctuated slightly in recent years, the overall geometry of the event distribution has been remarkably stable. This allows us to perform reliable resolution tests.

    We anticipate that phase data will be gathered at the temporary sites and used in global scale inversions for phase velocity maps as a function of period. One of many possible parameterizations for a global phase velocity map is an equal-area grid. The excellent ray coverage we expect for USArray justifies a grid for which the cell size is at least as small as 0.25 deg, between one third and one half of the station spacing. Figure 8 shows the results of a spike test (a point spread function) using phase data from one of the temporary arrays. The inversion has been done using an LSQR algorithm.

    A spike at 41.925 N/ 122.125 W with a 0.125o radius is fully recovered with extremely little smearing into adjacent cells observed. It is noteworthy that smearing in areas outside of the recording array is as small as within the array itself. Our tests also show that spikes outside of the recording array are also recovered in areas with adequate ray coverage (e.g. in the Pacific Ocean along the west coast), though LSQR requires more iterations to fully recover the anomaly amplitudes. Anomalies in areas with poor (or no data coverage) are smeared over adjacent cells (e.g. a spike southwest of Baja California) but not into the recording array. Note that erroneous mapping of structure outside of each temporary array can be easily avoided by augmenting the data gathered at the temporary array(s) with data from existing global permanent stations. One of the final products of the USArray experiment will be high-resolution phase velocity maps of the finest quality for the entire continental US and, of course, a detailed model of crustal and upper mantle structure. Note however that the data analysis and the modeling for structure can begin immediately after the data for the first temporary array have been gathered.

    According to a standard Backus-Gilbert analysis, Rayleigh waves in the period range 120-20s can resolve variations in shear velocity structure down to the transition zone though resolution degrades with depth. Phase data gathered for Rayleigh waves alone can constrain structure at a target depth of 25km to within ± 4 km (half width of resolution kernel), while structure at 400km will be smeared over ± 40 km. A combined analysis of Rayleigh and Love waves significantly decreases these uncertainties.

    A combined analysis of Rayleigh and Love waves provides important information on transverse isotropy (or radial anisotropy) within the upper mantle, perhaps as far down as into the transition zone. The resolution for Love waves is very good down to at least 400km and we anticipate that transverse isotropy can be mapped at an unprecedented scale. Such information is invaluable for constraining flow patterns in the upper mantle. For example, results from global modeling suggest that transverse isotropy is confined to the upper 200-250 km (though some studies find significant anisotropy in the transition zone). However, such data have been too sparse to accurately map transverse isotropy around the continental keels that might cause significant perturbations to the mantle flow pattern. Figure 7 reveals that the azimuthal coverage of ray paths is appropriate for studying the local azimuthal dependence of phase velocity (azimuthal anisotropy). Azimuthal anisotropy can easily be included in the modeling process of global phase velocity maps.

    There is a trade-off between azimuthal anisotropy and isotropic lateral heterogeneity when ray coverage is sparse. We anticipate that the dense station distribution and ray coverage will keep this trade-off at an extremely low level. Furthermore the extremely dense station spacing will give us the opportunity to perform other types of analysis. The small station spacing allows us to choose sub-arrays of approximately 4-8 stations to study the average properties within such an array. By simultaneously inverting for structural parameters and the incoming wave field, for each event, propagation effects due to structure outside the sub array and the earthquake source are eliminated. It is expected that the waveforms will be extremely coherent across the sub-array so allowing us to model the incoming wave field with only a few parameters (i.e. only small perturbations to a plane wave). The resulting average structure that will include azimuthal anisotropy is a genuine ‘local’ estimate of structure and can be used to generate continental-scale maps of anisotropy.

    Rayleigh waves in the period range considered here are also quite sensitive to variations in compressional velocities shallower than roughly 50 km and to density down to 250 km

    (though sensitivity to S-velocity is greatest). The data collected by USArray will let us investigate in detail the scaling relationships for the elastic parameters and density, as function of depth. This will allow us to conclusively discuss the compositional and thermal state of the lithosphere-asthenosphere system beneath the continental US.

    1.2.5 Active Source and Microearthquake Seismology

    High resolution of crustal structure and microseismic activity is largely a function of flexible array deployments and is therefore an important component for the envisioned multi-disciplinary experiments. The 2000 element high frequency array will be suitable for very dense active source profiling using vibrator and explosive sources, as well as for dense aftershock studies, and is therefore intended for a number of specific experiments. Resolution in active source experiments is a function of source density as well as receiver density.

    1.3 The Virtual Seismic Network. Current and emerging technologies allow us to deliver seismological data to an individual's desktop from a worldwide network of stations in near-real time. While many networks use dedicated communications links (landlines, radio, microwave) to transmit data from remote stations to the regional data centers, the technology to extend the Internet to remote locations is rapidly expanding and the Internet is increasingly becoming the primary mode of communication. The Virtual Seismic Network links data from diverse network sources through the Internet, to bring this data to researchers and students, to stimulate learning and enhance fundamental research in seismology.

    The real-time data component of the Virtual Seismic Network will be built around a heterogeneous mixture of existing regional seismic networks, the US National Seismic Network, USArray data, and PEPP stations. A schematic of the Virtual Seismic Network is shown in Figure x. From a heterogeneous mix of existing networks (which act as data sources, on the left) to a broad spectrum of potential users (on the right), the VSN will serve as a coordinating structure for real-time data distribution and provide tools for analysis, display and education. Once implemented, the Virtual Seismic Network (VSN) will, in essence, transform distributed sources of seismic data into a single, unified data source that end users with various skill levels can access from a single point of entry on the Internet. This involves development of a collection and integration system to homogenize the diverse data sources; an access and analysis system that provides users with real-time data and an infrastructure of software tools with which to process and display these data. The primary data flowing through the VSN will be time series of ground motion; however, the concept can be easily applied to any time series of environmental data and extended to include a variety of ancillary and derived data. In addition to providing an expanded resource for the research community, we will test and demonstrate the effectiveness of the VSN through the development of specific applications to support use of the VSN by K-12 students and teachers in their classrooms and by the general public in museum environments.

    The maintenance of full and readily accessible archives of continuous, quality controlled data is fundamental to the operation of any observational network This is especially true for seismology, where research may depend on access to observations of relatively rare events in particular locations. The primary function of the Virtual Seismic Network is to distribute and provide access to real-time data. In the distributed network concepts embodied in the VSN, we stress that the operational responsibility, mission-critical earthquake reporting and systematic collection, quality control and archiving of data for the various networks contributing to the VSN will continue to be an important responsibility of the individual networks. The archival function of the IRIS DMC adds value through the ability to mine the historical record of significant earthquakes and the observation of long term patterns of Earth behavior. In the VSN context, the DMC will provide access to special sets of historical data, acting as an additional data source, in addition to its function as a data archive. The creation of value-added products, related to significant earthquakes, will extend the resources of the VSN beyond real-time data. These products would assemble all available waveforms for selected large earthquakes (roughly one per day) in a customized data package, with supporting graphical displays, maps, seismic record sections, and ancillary information on the earthquake’s characteristics and effects. These products would then become available through the VSN and closely linked to the functionality of the VSN in educational settings. In addition to these event-based data sets being of special interest in educational programs, they will be a valuable resource for the research community.

    Starting at the end of 1998, a feasibility test was conducted for real-time data integration from multiple, disparate, seismic networks (Figure 2). Data from the IRIS GSN and Broadband Array, four US regional networks (Universities of Alaska, California at Berkeley and San Diego, and Nevada, Reno), and the Kyrgyzstan National Broadband Network were integrated into one common data processing system (Figure 2). A key component of this test was an extension of the software initially developed with IRIS funding as the central recording system for the PASSCAL Broadband Array. The "Antelope" software (from Boulder Real Time Technologies) extends this system and uses an "object ring buffer" (ORB) approach to solve a number of long-standing difficulties in seismological real-time recording:

    In addition to the primary data handling tasks, the Antelope system also provides preliminary data processing capabilities, including automatic picking of arrival times, event location and mapping, and the maintenance of a database of network operation. This test demonstrated that over 150 seismic stations from seven different primary data collection centers could be accessed through the Internet and processed in real-time using one SUN Ultra 60 workstation. The level of processing accomplished during this test included standard real-time functions that have been performed by individual network operators for more than 20 years: data assembly, automated phase picking, event location, and display of event location and magnitude information. As more scientists obtain access to real-time data streams we suspect strongly that entirely new functions will be devised to extract more information from these data in real-time.

    The Antelope ORB system will form the initial core of the front-end of the VSN as the Virtual Data Collection Center (VDCC, Figure 1). The function of the VDCC will be to :

    At present, the interfaces to the various networks include, in addition to Antelope, the Live Internet Seismic Server (LISS) developed at Albuquerque, the Near-Real-Time-System (NRTS) developed at UCSD, and the Earthworm system developed by the USGS and regional networks. In cooperation with the parent networks, these interfaces will be stabilized and improved and a standardized protocol will be established for the VDCC to communicate openly with the rest of the VSN.

    An Internet Data Distribution System. The distribution of real-time seismic data flowing into the VDCC systems to a distributed set of data nodes on the Internet and then delivery of that data to a potentially large number of end users is a critical component of the proposed VSN. This is termed the Internet Data Distribution System (IDDS, Figure 1). Specific elements of the IDDS are a reliable system to distribute and synchronize data nodes and a system to balance the load of users requesting data between the various nodes. We want to optimize access to these real-time data sets for people at a variety of locations. A system of IDDS nodes, initially based in the US but eventually distributed globally, will be used to optimize the distribution component of the VSN. This proposal requests support for the development of IDDS software and deployment of an initial set of six data nodes. We anticipate that significant cost matching will be available from participating universities and other organizations that may wish to extend the capability to include additional locations and other types of data.

    We have two potential models for the synchronization of nodes of the IDDS. First, an extension of the Antelope "ORB" utility described above that is currently being utilized to collect data from remote heterogeneous seismic networks. The scalability and suitability of this utility for large volumes of data and multiple distributed nodes will be evaluated. This can be viewed as the solution most compatible with the incoming data feeds of the VSN. A second solution is utilization of services to be developed within the access and analysis infrastructure described below. In this case as well, the suitability for the data volumes and load envisioned by the VSN will need to be evaluated. This solution can be viewed as most compatible with the outgoing data delivery component of the VSN. Our initial analysis of the advantages and disadvantages of each approach suggests that they are largely complementary in that the strengths of one system are often weaknesses of the other potential system. Thus, we are confident that an effective system for the distribution and synchronization of data from the VDCC to the IDDS nodes can be developed through utilization of a combination of these models.

    It is clear that a single machine can become saturated at some point as requests for VSN access and services grows, regardless of the data delivery system. We envision that users would only access the VSN through a generic domain name or URL, such as eol.org. The load balancing system would direct the connections to the most appropriate node for a given user. The function of this component is conceptually simple, but will require some development and testing to optimize performance. The Load Balancing System would take that request and hand it off to the IDDS node judged as the optimum server for the given client. The algorithm used to make this decision will require some development, as it must balance load on all the systems while simultaneously minimizing transport delays (e.g. by geographic distance or a more complex norm like the number of routers in a link). As this type of problem exists already for some large web servers, existing solutions will be investigated.

    1.4 Station Operations within USArray

    USArray is here interpreted in the narrow sense, corresponding to the elements of USArray that are anticipated to be funded through the NSF MRE process. USArray as defined in these terms has the following components and involves the following tasks:

    The transportable array

    The flexible array

    The permanent reference network

    Data from all components of the project will flow from the field deployments to the IRIS DMC and be available with minimal delays to the broad seismological community as well as for education and outreach other interested parties in a user-friendly format. The components of USArray as defined above will require a data archiving and distribution component that is not currently included within the scope of the MRE request. This need includes the acquisition of computer hardware and storage devices to archive, manage, and distribute the data from the USArray project.

    We envision three different modes of operation for the three components of USArray.

    The transportable array will be operated by a field crew dedicated to this effort. The 400 stations will be deployed in a grid with a station spacing of ~ 70 km for a period of ~18 months and then advance across the country in a rollalong fashion. The stations will be telemetered to a central site. The exact form of the telemetry will depend upon what is practical when the array is deployed. Whatever form of telemetry we adopt will be designed so that the waveform data enter the processing system via a TC/IP interface. Continuous data will be received at the USArray processing center from where it will be passed directly to the DMC. The USArray processing center will perform basic quality control on the data providing feedback to array operations in the field, create a catalog of events, window event data and submit data to the IRIS DMC for archiving. Real-time feeds of the data will be available to interested scientists from the DMC. Event data will be archived as soon as the events have been reviewed.

    The flexible pool of instruments is scheduled to operate in a mode similar to the current PASSCAL operations. That is, investigators propose special experiments to enhance the data gathered by the transportable array while it is located in a specific area. In this mode of operation the PI will furnish the bulk of the crew for operations just as is now done on PASSCAL experiments. PASSCAL will furnish training, logistical support and initial quality control and data formatting support. It is anticipated that most of the instruments in this pool will be deployed with some sort of telemetry. Data will be archived at the DMC as quickly as possible, and will be available through the DMC.

    The permanent reference network will be designed, acquired and deployed in cooperation with the USGS. Budget responsibilities for the operation and maintenance of the permanent stations will rest with the USGS. A statement outlining the design of the shared network and this mode of operation will be prepared and agreed to by NSF and the USGS. This cooperation assumes funding for the USGS ANSS project, or part thereof. The data will flow through real-time telemetry to the operator and then to the IRIS DMC as quickly as possible. The data from these stations will be integrated into the data flow from the transportable array.

     

     

    Appendix 2. USArray and the Deep Earth.

    The Earth's interior is a gigantic, evolving, heat engine with the mosaic of continental geology as one of its ultimate products. While USArray will provide substantial improvement in our understanding of the structure of the crust and lithosphere, these structures often result from processes related to mantle convection, with mantle plumes (hot spots), rifts, and subduction zones being the tangible flow trajectories. Clearly, a full understanding of continental geology and structure requires an understanding of the structure, dynamics, and evolution of the mantle. The USArray will lead to a substantial jump in our ability to resolve structure within important parts of the mantle. Through collaboration of seismologist with other deep Earth specialists, a much greater understanding of the dynamics and processes, which control mantle structure, differentiation, and ultimately continental evolution, will result from USArray.

    USArray will contribute in three ways to improving the resolution of deep Earth structure. First, from the transportable array, it will provide at a uniform 50-200 km resolution a three-dimensional "volumetric map" of seismic velocities and anisotropy beneath the United States down to at least 1,000 km depth. Second, within this volume of the mantle, the topography and sharpness of the principal jumps in seismic velocity will be mapped out beneath the continent with a lateral resolution of several tens of kilometers. Third, from the transportable array and judicious deployments of the flexible component, maps of core-mantle boundary structure at 10-100 km scales will result. Each of these improvements in mapping will provide fundamental constraints on the structure and dynamics within the regions of the Earth that contain the boundary layers of mantle convection.

    The upper mantle plays an extraordinarily important role in Earth dynamics and the three dimensional structure beneath continents has not been resolved sufficiently well to answer fundamental questions. It is quite clear that the upper mantle is anisotropic and this anisotropy probably results from the preferred orientation of minerals such as olivine, which is controlled by mantle flow. We do not know the details of the anisotropy and how it relates to changes in isotropic structure. We expect that the wealth of high quality broad-band data will stimulate the theoretical formulation of wave propagation in a geometrically complex, anisotropic media. In this way, the imaging of anisotropy can be turned into a powerful probe of three dimensional mantle flow. For example, the data provided by USArray should enable us to resolve the structure and deformation of the old Farallon slab and thereby determine how slabs sink across the transition zone into the deeper mantle. Furthermore, the thermodynamics of the two major jumps in seismic velocity at 410 km and 660 km depth, probably resulting from solid-solid phase transitions, are poorly known. The constraints from small, regional studies around the world are confusing and the absolute magnitude of the topography, the lateral scale-lengths of topographic undulations, the sharpness of the boundaries, and how the topography of the discontinuities varies with volumetric changes in seismic velocity are all poorly known. We know that the thermodynamics of these boundaries must have a first order influence on mantle dynamics, the mass flux between the upper and lower mantle, and the thermal/chemical evolution of the entire planet. With USArray we have the opportunity to determine the true nature of the mantle transition zone with unprecedented lateral resolution and how this relates to dynamic processes within the mantle, both upwellings and downwellings, and in turn how deeper dynamics is related to surface geology. This would be a spectacular advance for the Earth sciences.

    With USArray, it will be possible to empirically map out from the volumetric variations in seismic velocity and boundary deflections a "preassue-temperature-wavespeed phase diagram", the interpretation of which will require a vigorous interdisciplinary research effort between seismologists, mineral physicists, and geodynamicists. For the first time with USArray, we will have regionally, very dense coverage of a range of different seismic data (e.g. compressional and shear body waves as well as surface waves) which will greatly increase the utility of seismology for interpreting Earth structure and composition. Within seismology, the joint interpretation of P and S wave data (both for body and surface waves) is increasing in importance. Unfortunately, in most previous studies P and S wave images had different nominal resolution (e.g. shear wave maps from surface wave propagation and P wave maps from body wave travel times) which serious complicated, and often prohibited, the extraction of higher order information. Only for some local studies is high quality P and S wave information available at the same resolution. USArray will provide at high resolution a range of different data sets so that we will be able to constrain not only the spatial variations in wavespeed but also parameters which are much more meaningful for mineral physics: elastic moduli and their ratios (e.g., dlnVs/dlnVp, dlnVs/dlnVf, with Vf, the bulk sound speed, and Poisson's ratio). USArray will give us an opportunity to get 3D models of these parameters with a 50-200 km resolution to a depth of at least 1000 km. From mineral physics studies we have increasingly reliable estimates of the dependence of these parameters on temperature, pressure, composition, and mineralogy, so that we should be able to determine the lateral variations in composition and temperature with unprecedented accuracy and spatial resolution in a large fraction of Earth's upper mantle.

    Deeper in the Earth, the lower mantle and the core mantle boundary also play fundamental roles in the thermal and chemical evolution of the planet. USArray will be particularly useful in discerning the structure of the core mantle boundary because most of the teleseismic phases arriving in the US are from 70 to 120 degrees, ideal for detecting core mantle boundary structures. With the transportable and flexible components of USArray, we expect to map out structures in ~1000 km by 1000 km regions with resolutions of 10's of km beneath the central Pacific from events in the SW Pacific, beneath the Caribbean from events in South America, and beneath the Aleutians from events in the NW Pacific.

    The boundary between the core and mantle is a primary interface within the deep interior and has a fundamental influence on the magnetic field, the cooling of the planet, and volcanism at the Earth's surface. Although controversial, the CMB may even be the graveyard of hundreds of millions, if not billions, of years of subduction. Seismologists have revealed that the mantle side of this boundary is extraordinarily complex with a myriad of fine structure (e.g. < 10 km to a few 100 km's). Thermal and chemical heterogeneity, anisotropy, and melting within the lower mantle are all required to explain the observations and suggests that the lower boundary of the mantle is as complex as the mantle's top boundary (e.g. the continental crust and lithospheric mantle). The density jump at the core mantle boundary (CMB) is larger than that at the interface between hydrosphere and crust and the CMB boundary may be the site of long term and/or contemporaneous chemical differentiation within the mantle, in an analogous fashion to the way in which the Earth's surface layers are repositories of chemical and mechanical differentiation. Because of the extraordinary temperatures, which occur at this liquid metal-solid silicate interface, it could well be the most chemically reactive region in the planet. Consequently, understanding the origin of the fine structure of the lower mantle, in general, and ULVZs and D" in particular, is fundamental toward understanding how the solid Earth works as a globally interconnected system.

    In addition, core-mantle interactions is an extraordinarily exciting area and will likely grow in its relevance and significance as the seismological results of USArray start to come in. The detailed seismic studies mentioned above could potentially place important constraints on the mass exchange between the core and mantle. There are likely to be important dynamic interactions between the core and mantle as well that could be driven by lateral variations in the properties of the D" layer. Such lateral variations could have fundamental consequences for the generation of the geomagnetic field and its secular variation.

    With USArray, we expect not only spectacular advances in seismic imaging, but also advances brought about by the integration of seismology, mineral physics, petrology, and computational geodynamics. Indeed the pace of these activities within the Earth sciences has been picking up. There are outstanding recent examples of seismologists, mineral physicist, and geodynamicists working in various combinations; many of these collaborations have been made possible by special emphasis programs, such as the NSF CSEDI (Cooperative Studies of the Earth's Deep Interior) program. With the acquisition of high quality data with relatively fine spaced broad-band seismograph networks, such as USArray, the use of and motivation for advanced imaging and interpretation techniques dramatically improves. Likewise, the need for targeted interdisciplinary funding from the NSF increases.

     

    Spectacular advances are occurring in computational geodynamics and these have been driven in no small measure by the information technology revolution. The pace of these advances are likely to accelerate during the life of USArray. It is now becoming possible to simulate evolving continental plates within the whole mantle with fully realistic thermodynamics, chemistry, and mechanical properties. Geodynamic models allow two fundamental connections to be made which are vital for the interdisciplinary objectives of the USArray to be achieved. First, the models allow the results of mineral physics experiments and 'ab initio' results to be tied to tomographic images and the host of other information extracted from the seismic wave field. Second, geodynamic models allow the element of geologic time and the physics of mantle processes to be connected. Geodynamic models are time-dependent, evolving systems in which the laws of physics are self consistently incorporated. As an example, one fundamental use of dynamic models for USArray will be to explore in detail how the geologic evolution of the Farallon slab in the Jurassic is manifest in the present day seismic structure. Indeed, the numerical models allow one to tie geologic evolution, mineral physics, and seismology together.

    In summary, we anticipate that the USArray will produce spectacular advances in our knowledge of the structure, composition, dynamics, and evolution of the Earth's mantle. For the sake of completeness, the imaging activities of USArray can be organized into three different areas:

    1. Studies that depend on density of crossing ray paths from different directions, including all forms of seismic tomographic imaging. The primary targets will be the upper and mid mantle. We anticipate that lateral variations in seismic properties in the crust and mantle will be resolved at a scale of 50-100 km within the upper 1000 km of the mantle. Specific examples of such studies include:

     

    2. Site specific studies (including shear wave splitting, receiver functions, etc) at the lateral resolution of the transportable array (about 80 km, or less). With this data, we should be able to:

    3. Beam forming studies for deep Earth. With both the transportable and flexible components, we will be able to:

    Appendix 3: The Geologic Component of USArray. To translate the seismic images and geophysical measurements made with USArray into a better understanding of the structure and evolution of the North American continent will require input from a number of disciplines within the geological sciences.

    3.1 The need for multidisciplinary investigations. The seismic instrumentation employed in USArray will create an image of the current seismic velocity structure of the crust and upper mantle of North America. The seismic velocity of rock, however, is affected by a number of parameters including temperature, composition, mineralogy, shear fabric, and the presence of pore fluids. Several different approaches must be used to interpret the seismic velocity variations. These include:

    The first order translation of seismic velocities into rock properties is accomplished with input from the fields of petrology, rock mechanics and mineral physics. These techniques measure the elastic properties of the rocks and minerals that make up the Earth and provide the phase diagrams that show which mineral phases will exist at particular temperature and pressure conditions. Within the last decade, mineral physics has greatly expanded its ability to measure these parameters at the temperatures and pressures of Earth’s interior, rather than extrapolate from measurements at much lower temperature and pressure. This advance comes from developments in high-pressure apparatus, such as large-volume multi-anvil presses and the diamond anvil cell, coupled with extremely high intensity, synchrotron based, x-ray sources. Measurements of elastic properties are progressing at a rapid rate because of these developments, but much work remains to be done. The seismic images produced by USArray will further increase the need for high quality elastic parameters, and their pressure and temperature derivatives, in order to translate seismic velocity maps into rock type information.

    Field mapping of surface outcrop provides the most unambiguous information concerning the make up of the continent, but mapping obviously is only capable of examining the near surface. In areas where tectonism and erosion have brought deeper materials to the surface, petrologic and structural analysis provides a wealth of information concerning the history of the rocks and the conditions they experienced at depth. Field examination, thus, is a critical end member component of any study of continental structure and evolution. It provides an accurate and highly spatially resolved description of the continent’s surface. Some structures identified at the surface, for example, faults, folds and boundaries between rocks of different composition and/or history, can be extrapolated to depth. Confirmation of the accuracy of such extrapolations requires close coordination between the field analysis and the subsurface imaging that will be produced by the USArray geophysical instrument.

    Need paragraph on tectonic geomorphology and role of surface processes.

    Random fragments, xenoliths, of the deep crust and upper mantle, where they have been brought to the surface by explosive volcanism, allow additional constraints on the extrapolation of surface features to depth. Besides simple identification of what rock types are present in the deep crust and upper mantle, the technique of thermobarometry allows translation of the mineral composition of the xenoliths into temperature and depths of their capture by the rising magma. From this information, it is possible to construct the compositional stratigraphy and thermal profile of the crust and upper mantle. In some areas, as in the southwestern US, xenoliths can be found in eruptions of widely varied age. This makes it possible to examine the thermal and composition evolution of the crust and mantle as it is subjected to the changing magmatic/tectonic processes accompanying major continental deformation and magmatic events such as the Laramide orogeny and the formation of the Basin and Range province. Modern geochronological methods are capable of providing precise ages of formation and time-temperature histories for xenoliths. Such information is critical for understanding the degree to which surface and deep features are connected.

    Another way to examine the composition of the deep crust and upper mantle is through geochemical and isotopic study of magmas, now crystallized and exposed at the surface. In some ways, the geochemistry of igneous rocks provides a measure of deep crustal and upper mantle structure more akin to that provided by the seismic images. Actual samples of the crust and mantle, as in surface exposures or xenoliths, allow examination at the individual mineral scale, or with modern high-spatial resolution techniques, essentially down to the atomic scale. Seismic images are limited in spatial resolution by the wavelength of the seismic wave, of order hundreds of meters to kilometers. Similarly, magmas are derived by the accumulation of melts from "melting zones" many meters to kilometers in diameter. Thus, magmas provide something of an average composition for a volume of material of sufficient size to be resolved by USArray imaging. Furthermore, many issues of magma genesis and transport will be better understood with the imaging provided by USArray. For example:

    3.2 Time - The 4th Dimension. USArray will produce images of the current seismic and conductivity structure of North America along with a record of its current internal deformation. What the geophysical instrumentation of USArray will not provide is any information on how North America arrived at its present configuration. This information will derive primarily from geochronological studies of targets driven by the need to interpret USArray images. Modern geochronology can address many time scales, from the precise Holocene record now available through dating techniques that employ short-lived radioisotopes (for example, 10Be, 14C, 26Al, 36Cl, 210Pb, 226Ra and 230Th) to developments that allow uranium-lead ages to achieve precisions of +/- 1 million year on rocks as old as 4 billion years. At the short end of this time scale, geochronologic techniques can be directed at issues such as the record of movement along faults, the rate of uplift or subsidence in currently tectonically active settings, and the recurrence interval of active volcanoes. At the long end, geochronology can address issues of the rate of continent growth, identify terranes of differing age and history, define the timescale of deformation accompanying continental collision or rifting, and address whether seismic anisotropy in the crust and upper mantle is a response to modern strain fields, such as plate movement, or was frozen in from ancient deformation events.

    Geochronology is a technique of similar maturity to seismic imaging. Recent developments in broad-band sensors, array processing, and high-speed computing have vastly expanded the capabilities of seismic imaging and driven the call for a project like USArray. An advance of similar magnitude occurred in geochronology in the last decade. Ten to twenty years ago, geochronological studies of Precambrian rocks returned ages with uncertainties of 20 to 50 million years. A lot of geology was lost in this imprecision. For example, the whole history of the growth of the Rocky Mountains to initiation of extension in the Basin and Range to formation of the Snake River Plain would be considered a single event with this level of chronological resolution. With recent breakthroughs in analytical throughput and cost, and with the precision and range of applications currently available, resolution of most of the important events in the development of the continent is possible, and doing this at a continental scale is imaginable. For the study of more recent events, modern Ar-Ar dating can obtain ages precise to 10 — 100 Ka. Fifteen years ago, the eruption of the150,000 km3 of magma that forms the Columbia Plateau was believed to have occurred over a several million year interval because dating techniques could not resolve any shorter duration. We now know that most of the lava of this province was erupted in under a million years, which turns it from "just" a big volcano into an event of unprecedented magnitude in human history with dramatic potential consequences for the local, if not global, environment.

    The continental US was the site of many of the early geochronological studies of continents. Because of this, ages are available for much of the basement of the US, but many of these ages are of insufficient precision to interpret the growth history of North America to the resolution that will be required for proper interpretation of USArray images. One need only look to the remarkable success of Canadian geochronology where a systematic effort to obtain modern high-precision ages resulted in a history of the nearly 3 billion year evolution of the Canadian Proterozoic and Archean that rivals in detail the understanding of the evolution of the western US over just the last 60 million years. The quality of continent-scale imaging that will be provided by the USArray instrument deserves to be accompanied by a geochronological effort capable of resolving essentially all the major events involved in the assembly and break-up of North America. This is within the capability of modern geochronologic techniques.

    3.3 Heat flow, Potential Field, and Remotely sensed Data. (Heat flow, Robert N. Harris, University of Utah; Potential Field, and Remotely sensed Data, Randy Keller, University of Texas).

    The heat flow component of USArray. The thermal regime of the lithosphere greatly affects geologic processes and rock properties including: strength of the lithosphere and modes of deformation, depth distribution of earthquakes, concentration and orientation of stresses, metamorphism, electrical conductivity, density, and crustal magnetism. The thermal state of the lithosphere, in turn, is a manifestation of tectonism, magmatism and sublithospheric flow. As such, thermal processes are a fundamental component to understanding the tectonic evolution of the continent and geodynamics of the lithosphere. Incorporating heat flow data with the USArray is timely as the last sequence of papers on the thermal state of continental lithosphere culminated almost a decade ago [Crough and Thompson, 1976; Pollack and Chapman, 1977; Lachenbruch and Sass, 1984; Chapman and Rybach, 1985; Morgan and Gosnold, 1989]. Since then many more continental heat flow determinations have been made, and new ideas concerning the interpretation of heat flow data have been developed.

    An integral element to understanding continental deformation, evolution, and geodynamics involves careful consideration of surface heat flow and subsurface temperatures. The thermal state of the lithosphere is a function of the basal heat flow, crustal heat production, and tectonothermal components such as thermal convection, lithospheric deformation, magmatism, surface processes such as erosion and deposition, and groundwater flow. A fundamental question for each region analyzed by the USArray is, which of these processes provide significant contributions to observed surface heat flow? A proper evaluation of each of these potential effects is critical to distinguishing between them, important in their own right, and important in constraining thermal and mechanical models of the lithosphere. A second issue directly addressable by the USArray is the relationship between heat flow and continental elevation. In contrast to oceanic lithosphere where there is a clear relationship between heat flow and bathymetry no such relationship has been found for the continental lithosphere [Morgan and Gosnold, 1989]. Part of the problem lies in the relative structural and thermotectonic histories as well as the relative heterogeneity of the continental crust through variations in both crustal densities and thickness. A more complete description and better understanding of the continental crust, as proposed in the USArray will provide insight into this fundamental issue. We propose that for each region analyzed by the USArray, heat flow data (temperature as a function of depth and thermophysical rock properties) and analysis be integrated with high resolution seismic images, geologic, geochemical and deformation data to yield an improved understanding of the tectonic evolution, rheology, physical properties and lithospheric structure of the continental United States.

    Heat flow measurements can be obtained in one of two ways, either using "holes

    of opportunity" (mining exploration or dry holes drilled for water etc.) or by drilling new boreholes (200 m or deeper). After equilibration a thermistor is lowered down the borehole and records temperatures as a function of depth. These measurements are combined with measurements of the thermophysical rock properties (typically thermal conductivity and heat production). While hole of opportunity are not always ideally sited with heat flow measurements in mind they can be cost effective and provide new data points. Where justified, measurements in strategically drilled deeper boreholes generally provide superior data.

    References

    Chapman, D.S., and L. Rybach, 1985, Heat flow anomalies and their interpretation, J. Geodyn., 4, 3-37.

    Crough, S.T., and G.A. Thompson, 1976, Thermal model of continental lithosphere, J. Geophys. Res., 81, 451-459.

    Lachenbruch, A.H., and J.H. Sass, 1978. Models of an extending lithosphere and heat flow in the basin and range province, Geol. Soc. Am. Mem., 152, 209-250.

    Pollack, H.N., and D.S. Chapman, 1977, On the regional variation of heat flow, geotherms, and the thickness of the lithosphere, Tectonophysics, 38, 279-296.

    Morgan, P. and W.D. Gosnold, 1989, Heat flow and thermal regimes in the continental United States, Geol. Soc. Am. Mem., 172, 493-522.

    Gravity and Aeromagnetic Data. The publicly available regional gravity data set in the US in complete in most areas, and software to analyze these data is readily available and straightforward to use. Versions of these data are obtainable if one knows where to look. However, these data are sparse in some areas and data reduction procedures have been crude in areas with significant topographic relief. A small cooperative effort with the US Geological Survey would provide a very valuable data set that is usable be a large cross section of the geoscience community. This effort should encourage schools and government agencies across the country to compile data from sources such as these and local environmental studies, clean-up the data set with established criteria, apply consistent corrections, and store these data in a central location. The US is away behind most countries in that no national database of aeromagnetic data is available. While the country is covered with a variety of surveys, tying surveys flown at different times with different survey parameters (i.e. flight height, flight line spacing, etc.) is much more complicated than compiling gravity data. The USGS has an effort underway to create a database, but this is a daunting task given the variety of surveys.

    Remotely Sensed Data. A wide variety of remote sensing data are available across the US. These data are very useful but a new EOS sensor (ASTER - Advanced Spaceborne Thermal Emission and Reflection Radiometer) will provide powerful data tailor made for geologic applications. The objective of ASTER is to obtain high spatial resolution global, regional, and local targeted data in 14 channels from the visible through the thermal infrared wavelength regions, and these data will be sued to obtain detailed maps of surface temperature, emissivity, reflectance, and elevation. ASTER is the only high spatial resolution instrument on the EOS-AM1 platform. The ATER instrument consists of three separate instrument subsystems (Visible and Near Infrared (VNIR), Short-wave Infrared (SWIR), and Thermal Infrared (TIR)). The VNIR subsystem operates in three spectral bands at visible and near-IR wavelengths, with a resolution of 15 m. The SWIR subsystem operates in six spectral bands in the near-IR region through a single, nadir-pointing telescope that provides 30 m resolution. The TIR subsystem operates in five bands in the thermal infrared region using a single, fixed-position, nadir-looking telescope with a resolution of 90 m.

    Digital elevation models (DEM) will play a central role in many investigations accompanying USArray. A consistent and comprehensive DEM database will soon be available through the Shuttle Radar Topographic Mission (SRTM), a joint effort between NSAS and the National Imagery and Mapping Agency (NIMA). This mission uses C-band and X-band interferometric synthetic aperture radars (IFSARs) to acquire topographic data over 80% of the Earth's landmass between 60°N and 56°S during an 11 day shuttle mission. These data will be used to produce digital topographic maps products which meet Inerferometric Terrain Height Data (ITHD)-2 specifications (30 m x 30 m spatial sampling with 16 m absolute vertical height accuracy, < 10 m relative vertical height accuracy and < 20 m absolute horizontal circular accuracy). All accuracies are quoted at the 90% level, consistent with Nation Mapping Accuracy Standards.

     

    3.4 The Geologic Facility for USArray. The scientific need for a geological contribution to USArray is easily justified, but the geological "facility" to perform this work is less easily compartmentalized compared to the requirements for geophysical instrumentation. The geologic contribution to USArray will come from many different fields using many different techniques. As described below, much of the infrastructure for the geologic contribution is in place, distributed throughout US academic institutions. This "facility", however, is already subscribed in fulfilling the science mission for which it was originally established. To meet the additional demand imposed by USArray related projects would require that this distributed geologic facility be:

    Monitoring of volcanic activity is done through geodetic and seismic measurements, but also is aided by permanent gas (CO2, SO2) and hydrothermal activity monitoring facilities on active volcanoes. Such instrumentation is relatively inexpensive (of order $2,000 to $10,000 per site) until one includes the telemetry and data retrieval/management aspects of any continuously operating observatory. Telemetry and data management will be an intrinsic part of the geophysical components of USArray. Gas and hydrothermal activity monitoring could be easily, and relatively inexpensively, accommodated in USArray activities through the integration of GPS, gas and water temperature monitoring equipment into USArray geophysical stations installed on or near active volcanoes.

    Understanding the long-term (century to millennial) behavior of faults will come primarily from paleoseismological investigations. These studies include trenching across faults followed by detailed geochronology to reconstruct the past history of fault motion. Trenching is accomplished with earth-moving equipment of the type that easily, and relatively cheaply, can be contracted from a multitude of construction companies. Reconstructing the history of fault movement from the displaced layers observed in the trench walls is done primarily by 14C geochronology along with other cosmogenic nuclides supplemented by thermal luminescence and U-series isotopic measurements. NSF-EAR Instrumentation and Facilities currently supports one accelerator mass spectrometry (AMS) facility dedicated to 14C measurements, soon to be upgraded with an additional mass spectrometer, and another AMS laboratory specializing in other cosmogenic isotopes (10Be, 26Al, 36Cl, 129I). These facilities are heavily utilized, but have the capacity to provide an additional several hundred analyses per year. Thus, incorporation of paleoseismology into USArray is best done by ensuring an adequate science budget that will allow the contracting of both the trenching efforts and the geochronologic analyses.

    To investigate the longer term behavior of both faults and volcanoes, as well as the general geologic evolution of North America, requires detailed field study coupled with chemical and isotopic analysis and high-precision geochronology and thermochronology, primarily (U-Th)/He, Ar-Ar and U-Pb dating of surface rocks and xenoliths. Access can be one of the largest impediments to fieldwork. Given the needs of USArray in permitting and site access, ready field access could be provided by the USArray geophysical effort at little additional cost.

    Facilities for rock chemical and isotopic analysis and geochronology are available in many US academic institutions. These too are heavily utilized, but are capable of redirecting their effort towards USArray related projects if they choose to do so. Traditionally, most of these facilities are operated by individual faculty at universities and involve only limited technical support. Consequently, the limiting step in most of these operations is not the instrumentation, but is the personnel time necessary for sample preparation. In the example afforded by the Continental Dynamics Program, these laboratories usually are involved in multidisciplinary projects when the person in charge of the laboratory, and/or their graduate students, are personally involved in the study. Some geochemical and geochronological laboratories also operate on a charge per analysis basis, but the capacity here is limited by the limited technical support available to most such laboratories. Consequently, per analysis recharge rates are high, and throughput well below that which could be achieved if additional technical support were available. If the USArray demand for geochemical and geochronological analyses cannot be met by redirection of the scientific interests of existing laboratories, the efficiency of existing laboratories could be increased. This could be accomplished by providing funding to those laboratories having appropriate instrumentation to allow efficient contract analyses if they had the dedicated technical personnel required to carry out this task.

    Appendix 4: The GPS Component of USArray. USArray will give us a detailed picture of continental architecture. This architecture has presumably evolved over many millions or even billions of years, and can be thought of as the integrated effect of tectonic processes over that time. However, in much of the western US, it is likely that the youngest activity, operating over the last few million years to the last few tens of millions of years, is responsible for much of the architecture that we can observe today. This activity also provides a modern "snapshot" of at least some of the processes that presumably have operated over most of geologic time. If we hope to understand the existing architecture, and if we hope to look past the young overprint to earlier processes, it will be necessary to have the clearest possible description and understanding of the youngest processes. A detailed, continent-wide surface velocity field from GPS can contribute significantly to this understanding. Recently published velocity fields for the western US (Bennet et al., 1999; Thatcher et al., 1999) illustrate what is possible, but many gaps exist, and current stations, especially in the eastern US, are not optimized for high accuracy. GPS has proven valuable for investigating mid-continent seismicity at the New Madrid seismic zone (Liu et al., 1993; Newman et al., 1999) and plate rigidity (Dixon et al., 1996), but current results are strongly limited by station coverage and accuracy. In large regions of the Basin and Range province and the northwestern US, GPS coverage is still relatively sparse, and the active tectonics here demand much better coverage. USArray can not only improve the sampling density in regions where few stations currently exist, but also improve accuracy. Strict configuration control is essential to achieve highly accurate GPS velocities in a consistent reference frame. USArray can provide the necessary framework, analogous to what SCIGN and BARD have done for southern and northern California, respectively, and give us a much better snapshot of the modern tectonic processes that play a role in continental architecture.

    A continent-wide GPS network has applications to both the USArray and Plate Boundary Observatory components of the "EarthScope" initiative, as outlined below. As currently configured, USArray will include 16 high quality permanent GPS stations, to be collocated with the permanent seismic stations, and an unknown number of additional stations that could be acquired in conjunction with the 400 station Transportable Array of seismic instruments ( "Bigfoot") as it moves across the country, eventually encompassing 2,000 sites. Here we argue that there should also be a GPS component to the Transportable Array, and that it should be an integral part of the seismic instrumentation. In this way we can eventually construct a high quality, nationwide GPS network. Since the GPS and seismic instrumentation would share infrastructure, the cost of this approach is much cheaper than the cost of constructing such a GPS network from scratch.

    At least 65% of North America's area has potential "signal" for a continent-wide array of high precision GPS stations. Even the remaining "stable" area is important for GPS observations, partly because such stations help define a stable North American reference frame for the velocity field, and also because it is possible that signals will be detected here that are beyond our current understanding of displacements and stresses for stable cratonic regions (recall the seismic sequence at New Madrid in 1811-1812, for which we lack a viable explanation). Here we define tectonic signal for a GPS network as any horizontal or vertical motion exceeding 0.5 mm/yr., a velocity threshold that is detectable with GPS with measurements spanning about 5 years (horizontal components) and about 10 years (vertical component).

    4.1 The scientific applications for a continental GPS array include:

    4.2 Scope, Cost and Logistical Considerations. To provide a cost estimate for adding a continental GPS network to USArray we assume that the GPS network will eventually consist of 1000 sites, i.e., 50% of the 2000 Transportable Array seismic stations. Of these, 700 GPS stations would be in the western third of the US (100% of the USArray stations west of the Colorado Plateau would have GPS measurements), and 300 would be in the eastern two thirds of the US (roughly 25% of the seismic stations east of the Colorado plateau).

    The major cost in establishing any GPS network is the cost to support trained personnel to locate a number of candidate sites, choose among a subset of possibilities, and arrange for the necessary permits from government agencies or approvals from landowners. When these manpower costs are included, typical per site costs for a GPS installation can exceed $100K. Other significant costs include designing, purchasing and constructing security fencing, power and data telemetry. Given the infrastructure to support the seismic installations of USArray includes permitting, power, data transmission, and site security, the marginal cost of adding GPS to the Transportable Array is low, consisting mainly of the receiver hardware and the monument. Board level GPS receivers are now available for less than $6K, and could be incorporated directly into the seismic instrumentation, considerably easing deployment logistics. In fact, existing IRIS broadband systems already include a GPS receiver for timing. In effect we are only proposing to upgrade this board, and add a geodetic quality antenna. One cost unique to GPS is the monument. Monument costs can range from less than $10.00 (a stainless steel pin set in bedrock) to $500-$1,000 for standard (~1 m3) concrete and rebar monuments, to $10,000-20,000 for deeply drilled, ultra-stable monuments. The latter may not be necessary in cases of good hard rock outcrop, given that noise sources other than localized ground motion currently dominate the GPS error budget if competent rock is available. In alluvium, however, careful attention will have to be paid to monument design.

    The above cost considerations suggest that for a modest additional investment, USArray seismic systems could be outfitted with GPS. Assuming that the transportable array will eventually include 400 instruments and will eventually cover 2000 sites, of which we will monument half, and assuming that the GPS component is fully integrated into the hardware design and operations, for example sharing site security, power and data telemetry, then the marginal cost of GPS at these USArray sites is mainly the cost of receiver/antenna hardware (total 400 systems) plus monuments (total 1000). An estimate of the hardware and monument costs for 400 systems deployed in sequence at 1,000 USArray sites is ~$8 million. These costs could potential be reduced through an early joint development effort between IRIS and UNAVCO and there could be significant operational savings if all site hardware and installation/take down procedures are completely standardized. The cost could be higher if the community decides to install more expensive monuments. If we arbitrarily assume that 10% of the sites (total 100) will require such monumentation, then the added cost is $1.5M. Thus the total cost of the GPS augmentation to USArray would be $9.5 million.

    We suggest that UNAVCO establish a design team to work with IRIS and produce a high quality, moderate cost monument design for use in the transportable component of USArray as the "standard" design, to be used at the majority of the sites. Specific design criteria should include:

    A 1000 station GPS network across North America coinciding with the 2000 station transportable seismic array could be a cost effective way to identify high-quality sites for permanent GPS installations as part of PBO. Sites not occupied with PBO instrumentation could be resurveyed in campaign mode. Site density would be highest in the western US, where all of the USArray Transportable Array sites would be occupied, giving about a 70 km station spacing. GPS should be completely integrated into the transportable array seismic system by UNAVCO and IRIS. GPS data should be archived and openly available in near-real time. Site occupation for GPS measurements should be identical with seismic instruments (1.5 years), to minimize logistical difficulties. Re-occupation of the GPS sites should be performed 3-5 years later in a campaign consisting of short (1-2 month) observations at each site to improve velocity accuracy (accuracy is largely a function of total observation time). The cost of subsequent occupation of the entire network in campaign mode is estimated at ~$1.8 million.

     

     

    Appendix 5: The Magnetotelluric Component of USArray. (Steve Park, University of California, Davis). The value of complementary magnetotelluric (MT) surveys to multidisciplinary studies of active processes forming and reshaping continental crust has been demonstrated

    repeatedly over the past decade. In one of the earliest joint seismic-MT studies, subducting sediments were mapped beneath the Pacific Northwest (Jones, 1987). A

    multidisciplinary study of the southern Sierra Nevada has shown that these mountains are rootless (Wernicke et al., 1995), and it was a combination of MT data (Park et

    al., 1996) and a seismic array study (Jones et al., 1998) that showed the mountains were supported instead by partial melt in the upper mantle. A combination of seismic

    and MT data (Chen et al., 1996) has shown that significant partial melt exists in the lower crust in Tibet, leading to new concepts about the deformation of the crust at the

    collision between the Indian and Eurasian plates (Nelson et al., 1996). In studies along the San Andreas fault zone, Mackie et al. (1997) and Unsworth et al., (1997) have

    mapped the distribution of fluids and inferred zones of strength and weakness. Most recently, a MT study (Park and Mackie, 1997) has mapped the distribution of fluids in

    actively deforming crust, leading to new views about the process of metamorphism in the Himalaya.

    5.1 The scientific applications for continental scale magnetotelluric images. The MT method is most sensitive to conductors in the crust and upper mantle. Silicate minerals at subsolidus temperatures in the crust are very resistive, so natural electrical

    currents are attracted to regions of low resistivity. These conductive regions can be caused by solid phases such as graphite and metallic sulfides or oxides, aqueous fluids

    (especially brines), and partially melt (Jones, 1992). While there is a weak dependence of electrical resistivity in the crust on temperature, the variability caused by changes

    in state and fluid content is much larger than the general decrease caused by temperature. A common requirement in all of these causes is that the conductive fraction must

    be interconnected, thus leading also to inferences about fluid migration paths. In the mantle, the apparently simpler mineralogy leads to fewer potential causes of lower

    resistivities. Resistivities of silicate minerals in the mantle decrease systematically with temperature, allowing estimates of temperature from MT soundings. Additionally,

    aqueous fluids do not appear to wet the surfaces of ultramafic minerals (Watson and Brenan, 1987), so conductive fluids in the mantle are confined to basaltic melts.

    Recent MT studies have shown that the 440 km and 660 km seismic discontinuities are also resistivity boundaries (Neal, 1998).

    A common thread in most of the field investigations mentioned above is the use of seismic arrays and MT soundings to delineate crustal and upper mantle structure. Seismic properties such as attenuation and compressional and shear wave velocities can be coupled with electrical resistivity to allow discrimination of the type and state of crust. For example, the conditions under which partial melting can occur can be ascertained by the joint use of velocity and resistivity. Laboratory data (Olhoeft, 1981) and petrologic considerations (Wannamaker, 1986) have shown that partial melting of granitic crust in water-undersaturated conditions leads to partial melt that can be almost as resistive as the surrounding rock, while melting in saturated conditions leads to conductive melts. In both cases, partial melt will lower seismic velocities. In rocks with aqueous, interconnected brines, small fractions of a percent fluid content can lower the electrical resistivity but leave the velocity unchanged. Thus, a rock with low resistivity and normal velocities could have a small fraction of aqueous fluids (Table 1). At higher fluid fractions, both resistivity and velocity would be lowered. Neither method alone can discriminate between possibilities, but the combination of both can help identify the state of the crust.

    Table 1. State of Continental Crust from Resistivity and Velocity

      High Vp, Vs Low Vp, Vs
    High resistivity No melt Water-undersaturated melt
    Low resistivity Aqueous fluids (<< 1%)

    Water-saturated melts

    (or aqueous fluids)

    A fundamental concern with the MT method is that its responses at long periods are sensitive both to deep and to distant structure. Because of the resistivity contrast between the air and rock, the transmission of MT fields into the Earth is via diffusion and not wave propagation. Short periods penetrate only to shallow depths and are affected primarily by shallow/local structure, but long periods sense both shallow and deep structure. Because the fields are diffusive, regional structure can affect measurements at a given location. For example, Park et al. (1996) showed that the electrical influence of the conductive Pacific Ocean was seen as far inland as 200 km away in the Sierra Nevada. More recently, Mackie et al. (1996) showed that the electrical structure of the California Transverse Ranges influenced electric fields over 200 km away in Death Valley. Interpretation of MT data is truly a 3-D problem but few surveys have the regional distribution of stations to permit development of a regional 3-D model. An additional concern is that no practical 3-D inversion programs currently exist, but progress is being made on their development (Newman and Alumbaugh, 1997). The most common approach to interpreting data in this regional 3-D environment is to select profiles across approximately 2-D structures and then use only those portions of the data that are validly represented with a 2-D model along the profile (e.g., Wannamaker et al., 1984; Park and Mackie, 1997). Coupling MT soundings with the transportable array would provide this needed 3-D regional MT structure and permit a quantum leap in our ability to interpret data. Reinterpretation of existing surveys in this regional context could reveal structures missed in the earlier, 2-D models, and new surveys would be planned with the regional network in mind.

    An outstanding scientific issue is the controversy over the existence of fluids in the old continental crust. Many MT surveys (e.g., Jones, 1992 for summary) report the existence of moderately conductive lower continental crust, but petrological considerations appear require a dry lower crust (Yardley and Valley, 1997). Attributing the conductivity to fluids (either brines or melts) in active margins is reasonable, but how can this phenomenon be explained in old continental crust? Possible explanations have included graphite, sulfides, and oxides, but these are controversial because they must be connected phases (Yardley and Valley, 1997). The habits of these minerals, as well are their conditions of formation, seem to preclude interconnection. Similarly, fluids would have presumably been bound into minerals formed during retrograde metamorphic reactions. If free fluids are present, how have they been trapped in the lower crust? One criticism of the conclusion that the lower crust is conductive is that MT surveys are most often done to delineate anomalous crust (Wannamaker, P.E., Comment on "The petrologic case for a dry lower crust" by Yardley and Valley, submitted to J. Geophys. Res., 1998). Perhaps all we can conclude is that anomalous crust is conductive and we do not know the properties of normal crust. A second advantage of coupling MT soundings with the "Bigfoot" array would be a systematic sampling (albeit sparse) of the entire continental US, which is underlain in many areas by old, stable continental crust. While such a survey would not resolve the cause of lower crustal conductivity, it would determine if low resistivity is indeed the norm for old, stable crust. In any case, comparison of regions with high and low resistivities could help understand the causes of lower crustal conductivity.

    Magnetotelluric Instrumentation: Standard commercial broadband MT systems span a period range from 0.001-1000 s and are used primarily for studies of the upper and lower crust. A useful measure of the depth of penetration is the skin depth (SD = SQRT(rho*T), where rho is the resistivity and T is the period). For an average resistivity of the crust of 100 ohm-m, this system will penetrate to depths of 150 km or less. These systems use induction coils for which the output voltage is inversely proportional to the period of the signal. At periods beyond 1000 s, the coil response often drops below the noise level of the sensor. For studies of the mantle, long period MT systems using fluxgate magnetometers are employed. These magnetometers are less sensitive than induction coils at the shorter periods, but are capable of recording the larger signals at the longer periods. The electric field sensors in both systems are similar.

    The principal difference between the broadband and long period MT systems is the length of time for which a station is occupied. Errors of less than 5% in apparent resistivity and 1-2 degrees in phase are achievable at periods of less than 1000 s with a broadband system in 24-48 hours of recording. Similar errors are achievable with the long period system at periods of 10000-30000 s with 2-4 weeks of recording. Broadband systems are thus designed to be operator-controlled via a computer while long period systems are normally built around a data logger. In comparison to seismic instrumentation, the broadband systems would be like refraction or reflection systems and the long period systems are like seismic stations. At still longer periods capable of sounding into the middle mantle, permanent installations which record data over a period of 1-2 years are utilized. Most often, these use telephone lines as electric field dipoles and fluxgate magnetometers for magnetic measurements. Siting of these stations is very critical because electromagnetically noisy sites cannot be used.

    There may be little or no advantage to routine recording continuously over a long period of time at stations, unlike the situation with seismic arrays. In a study being conducted at Parkfield, California, scientists have shown that errors of 1% in apparent resistivity and 1 degree in phase are achievable with continuous recording at broadband periods. However, this is still too large an uncertainty to make the MT method useful for monitoring tectonic changes in the crust (Park et al., 1993; Park, 1996). Laboratory and field measurements both show that resistivity changes prior to earthquakes are likely to be no more than 1-2% at best, and the effect of these material property changes would be diluted by the much larger volume of unchanging crust. Nonetheless, it could be useful to allow for the possibility that some continuous recording would be done in tectonically active areas.

    A common technique for reducing the effects of noise in an MT sounding is to use cross-correlation between two sites recording simultaneously (e.g., the remote reference method). Studies of electromagnetic noise in developed countries shows that the site separation may have to be 100-200 km in urban areas. One advantage of a few continuously recording stations across the country would be the availability of good base stations for surveys nationwide. In addition, these stations could be used for studies of the middle mantle.

    Proposed is a 3-tiered system of MT stations, mirroring the seismic component of the USArray. The first tier would consist of permanent stations on a 500 km station spacing (12-15 sites) for remote reference stations and studies of the mantle. Each station would consist of a fluxgate magnetometer and long (1-10 km) electric field dipoles sampled once per second. Initial installation costs would be approximately $15,000 per site and annual costs of $3,000 per site could be expected. Broadband instruments could be collocated temporarily at the permanent stations when needed for short period remote referencing.

    The second tier would travel with the transportable array and consist of both long period and broadband MT systems. As each new seismic station was established, a broadband MT station would be installed and recorded for a period of 1 month. Each long period system costs approximately $20,000, and a pool of 20 instruments would be sufficient to record MT responses at each station in the transportable array. Not every station in the transportable array would be a suitable candidate for MT measurements; stations would have to be located at least 1 km from the nearest source of AC power and road vibrations. Finally, broadband MT fields would be recorded at each station for period of 4-5 hours. A pool of 3-5 broadband stations would be adequate for this purpose. Each broadband station costs approximately $75,000. With this complementary coverage of broadband and long period recording, sampling of the MT response at all depths from the shallow crust (< 1 km) to the upper mantle (>200 km) would be possible. These MT soundings, at spacings of 70 km, would provide an ideal framework within which to assess regional 3-D MT fields.

    The third tier would consist of broadband and long period measurements at the sites of the flexible array. Survey plans would be driven by the scientific goals of the regional workshops; it is not possible to divine these in advance. The same instruments used for the second tier of studies could be used for this level of study, however. If each transportable array is occupied for a period of one year and consists of 200 stations, a pool of 20-30 long period instruments would be available for flexible array studies. A pool of 4 broadband systems might suffice for flexible array studies. There is also the possibility that existing EMSOC equipment (2.5 broadband systems and 14 long period instruments) would be available for the flexible array studies.

    References

    Chen, L., J.R. Booker, A.G. Jones, N. Wu, M.J. Unsworth, W. Wei, and H. Tan, 1996, Electrically conductive crust in southern Tibet from INDEPTH magnetotelluric

    surveying, Science, 274, 1694-1695.

    Jones, A.G., 1992, Electrical conductivity of the continental lower crust, in Fountain, D.M., R.J. Arculus, and R.W. Kya (eds.), Continental Lower Crust, Elsevier, New

    York, 81-143.

    Jones, A.G., 1987, MT and reflection: An essential combination, Geophys. J. Roy. Astr. Soc., 89, 7-18.

    Jones, C.H. and R.A. Phinney, 1998, Seismic structure of the lithosphere from teleseismic converted arrivals observed at small arrays in the southern Sierra Nevada and

    vicinity, California, J. Geophys. Res., 103, 10065-10090.

    Mackie, R.L., T.R. Madden, and S.K. Park, 1996, A three-dimensional magnetotelluric investigation of the California Basin and Range, J. Geophys. Res., 101,

    16221-16239.

    Mackie, R.L., D.L. Livelybrooks, T.R. Madden, and J.C. Larsen, 1997, A magnetotelluric investigation of the San Andreas fault at Carrizo Plain, California, GRL, 24,

    1847-1850.

    Neal, S.L. 1998, Variations in the electrical conductivity of the upper mantle beneath North America and the Pacific Ocean. MS Thesis, Indiana University, Bloomingdale.

    Nelson, K.D., W. Zhao, L.D. Brown, J. Kuo, J. Che, and others, 1996, Partially molten middle crust beneath southern Tibet: Synthesis of project INDEPTH results,

    Science, 274, 1684-1687.

    Newman, G.A., and D.L. Alumbaugh, 1997, Three-dimensional massively parallel electromagnetic inversion, 1. Theory, Geophys. J. Intl., 128, 345-354.

    Olhoeft, G.R., 1981, Electrical properties of granite with implications for the lower crust, J. Geophys. Res., 86, 931-936.

    Park, S.K., Seismoelectromagnetic effects, Surveys in Geophysics, 17, 493-516, 1996.

    Park, S.K., B. Hirasuna, G.R. Jiracek, and C.L. Kinn, 1996, Magnetotelluric evidence of lithospheric mantle thinning beneath the southern Sierra Nevada, J. Geophys.

    Res., 101, 16241-16255.

    Park, S.K., M. J. S. Johnston, T. R. Madden, F. D. Morgan, and H. F. Morrison, 1993. Electromagnetic precursors to earthquakes in the ULF band: A review of

    observations and mechanisms. Reviews of Geophysics, 31, 2, pp. 117-143.

    Park, S.K., and R.L. Mackie, 1997, Crustal structure at Nanga Parbat, northern Pakistan, from magnetotelluric soundings, Geophys. Res. Lttrs., 19, 2415-2418.

    Unsworth, M.J., P.E. Malin, G.D. Egbert, and J.R. Booker, 1997, Internal structure of the San Andreas fault at Parkfield, California, Geol., 25, 359-362.

    Wannamaker, P.E., et al, 1984, Magnetotelluric responses of three-dimensional bodies in layered Earth's, Geophys., 49, 1517-1533.

    Wannamaker, P.E., 1986, Electrical conductivity of water-undersaturated crustal melting, J. Geophys. Res., 91, 6321-6327.

    Watson, E.B. and J.M. Brenan, 1987, Fluids in the lithosphere, 1. Experimentally-determined wetting characteristics of CO2-H20 fluids and their implications for fluid

    transport, host-rock physical properties, and fluid inclusion formation, EPSL, 85, 497-515.

    Wernicke, B., R. Clayton, M. Ducea, C.H. Jones, S. Park, S. Ruppert, J. Saleeby, J.K. Snow, L. Squires, M. Fliedner, G. Jiracek, R.Keller, S. Klemperer, J.

    Luetgert, P. Malin, K. Miller, W. Mooney, H. Oliver, and R. Phinney, 1996, Origin of high mountains in the continents: the southern Sierra Nevada, Science, 271,

    190-193.

    Yardley, B.W.D. and J.W. Valley, 1997, The petrologic case for a dry lower crust, J. Geophys. Res., 102, 12173-12185.

     

     

    Appendix 6: Education and Outreach (Catherine Johnson, Program Director, IRIS Education and Outreach Program; Trail of Time Concept introduced by Karl Karlstrom and Michael Williams, University of New Mexico) While the primary motivation for USArray is grounded in fundamental advances in scientific discovery, the initiative provides a spectacular opportunity for a comprehensive education and outreach program that will extend across the country and continue throughout and beyond the lifetime of USArray. USArray capitalizes on the public's natural interest and excitement in earthquakes and Earth science and can make these subjects relevant on a region by region basis as the array is deployed across the country. The USArray initiative offers many possibilities for outreach and education to reach the general public, K-12 students and teachers, and Earth science students and professionals at all levels. We anticipate a comprehensive program that tries to convey both the new scientific results that emerge from the USArray national scientific effort, and perhaps as importantly, the nature of our scientific method. This effort will be strengthened as USArray moves from region to region over a 10 year period and highlights both important regional questions and an emerging (and changing) continental scale picture. There is a great potential for science PIs to work closely with trained education and outreach people to effectively translate and communicate both scientific results and the scientific method to all audiences.

    The span of geologic time is one of the most difficult concepts for geologists and non-geologists to assimilate. Yet an understanding of this concept seems increasingly essential if people are to understand and make decisions about the complex environmental problems such as global warming and climate change, air and water pollution, hazardous waste storage to mention just a few. Each of these issues involve time spans that can be difficult to relate to the human scale but gain new relevance on a geologic scale. The "Fourth Dimension" of the USArray effort is also about time, and how we can combine temporal understanding of tectonic history with the new images of the 3-D structure of our continent. This combination of elements in USArray, and its national scope, may provide an exceptional opportunity for Earth scientists to communicate to the general public. This is especially important in view of the continued need to influence policy decisions about natural resources, geologic hazards, and science education.

    USArray provides an excellent opportunity to increase public awareness of science, Earth science and seismology, improve communications between K-12 schools, community colleges and research institutions, and improve the image of Earth science and of scientists. Not only does USArray provide an educational forum across the country in terms of regional geologic structures, natural resources, hazards, and society's relationship with the Earth and the environment, it also provides the opportunity to engage the public in the scientific process by involving them in a research enterprise. The USArray initiative can reach national audiences through PBS/Discovery Channel programs, an "Earth Minute" on NPR, and regular feeds to Weather Channel. The latter could include earthquake reports, real-time data feeds, and an "Earthquake of the Day" and could highlight geologic structures and hazards that are unique to the region currently occupied by the array.

    As a result of the National Science Education Standards and state standards, many K-12 teachers are having to teach Earth science for the first time. USArray can help get solid content into the curriculum through a real experiment. Students can learn science by doing science. In order to maximize the potential for USArray to engage broad participation across the country we will need to develop both overview materials that apply to the initiative as a whole in addition to materials that highlight hazards, natural resources, and the geologic structures and history that are specific to each region covered by the array. The USArray Education and Outreach team will work with local teachers, scientists, colleges, museums, and state geological surveys to develop hard-copy and web-based materials that will be associated with USArray. A modular museum display will be developed with some components of the exhibit installed before the array is in place in the specific region. A real-time display will be added while the array is in place with a "grand opening" immediately after the array is installed.

    To reach the widest audiences, there will need to be many different types and styles of outreach. Ideally outreach developments would include multilevel use and dynamic displays. One potential theme that could have applications across the country is the concept of the "Trail of Time. At the Grand Canyon, a walking trail on the South Rim is under development that is scaled such that 1 meter is approximately 1 million years of Earth history. A walk through the Canyon’s two-billion year history will be 2km long, with large, spectacular, rocks interspersed at appropriate spots along the trail. The rocks, the gaps, and the displays along the trail can communicate the concept of geologic time and how geologists use the incomplete record preserved in rocks to reconstruct Earth history. At the beginning and end of the trail are possibilities for more specialized intersecting trails such as "What is a Million Trail" (to give people the idea of how big this number is), "Trail of Humanity" (People in the Grand Canyon region), and "Canyon Cutting Trail" (a history of conflicting and emerging scientific ideas for cutting of the Canyon).

    There is an opportunity to develop regional "Trails" that would emerge as USArray moves across the country. The trail concept can be modified as appropriate for each of the regional efforts and could be housed at Universities (the New Mexico Trail of Time at the University of New Mexico for example), at National Parks (The Rocky Mountain Trail of Time at Rocky Mountain National Park, The Hotspot Trail at Yellowstone, The Appalachian Trail of Time at the Great Smokey Mountains, etc.), at Museums (the Smithsonian Trail of Time on the Mall), or at USGS or State Survey offices. Another goal of the "Trail of Time" will be to present some of the basic geologic processes that are shaping the Earth today. A "Cascadia Trail of Time" might discuss volcanic processes and subduction-zone earthquakes. A "New Madrid Trail of Time" might discuss intraplate deformation and structure. A San Andreas trail might specialize in Earthquake processes, prediction, and mitigation of damage.

    6.1 Development of Educational Products. As USArray moves across the country it will be coordinated with a comprehensive educational and outreach program highlighting both overarching and regional Earth science issues (hazards, structures, resources) and links between Earth science and society. Specific initiatives will include development of Virtual Seismic Network (VSN; see Appendix 6.), software for use by educators in K-16 educational programs, museum displays, public and commercial media programs, teacher workshops, and educational materials and activities linked to array data and Earth models available over the Internet.

    Educational products of USArray

    • Virtual Seismic Network
  • Museum display
    		•
  • General educational materials (posters, 1-page synopsis, activity folders for teachers and museums)
    		•
  • Region specific educational materials highlighting regional Earth science issues (hazards, structures, resources, links between Earth science and society)
    		•
  • "Earth Minute" on NPR
    		•
  • USArray Web site (detailed below)
    		•
    • Broadcast quality educational videos (~ 15 minute length video on US Array available for teachers, media, and general information and a shorter 5 minute version for use in a museum setting and/or by media
  • Press packet with fact sheets, information from previous deployment, information on PR/education programs and opportunities associated with USArray.
    		•
  • PBS/Discovery Channel programs
    		•
  • Feed to Weather Channel
    		•

     

    6.2 USArray web site will need to serve multiple purposes: internal planning and information services for participating scientists and engineers, public outreach and information dissemination, access to data for scientists, access to data for non-specialists, and dissemination of educational resources. The web site will thus require collaborations between scientists with regional expertise and the IRIS DMC, PASSCAL, GSN and the E&O program for content. As the web site will be a major outreach vehicle and press resource for US Array, providing a consistent, user friendly interface is critical.

    USArray Website

    • General info on USArray
  • Earthquake maps
    		•
  • Time history of deployment, high profile results, science questions
    		•
  • Access to Data (web tools and software for downloading/viewing data in real-time
    		•
  • Earthquakes of interest recorded by the array
    		•
  • Current earthquakes/teachable moments
    		•
  • Tools for data download and visualization
    		•
  • Information on workshops for teachers, internship opportunities
    		•
  • Site info (photos, technical info etc)
    		•
  • Results from prior deployments
    		•
  • Links to data, data return statistics
    		•
  • Educational materials and resources
    		•
  • Links to local sites
    		•
  • Acknowledgements and contacts
    		•

    6.3 Pre-installation regional education and outreach efforts in given area will require public relations and related education efforts to establish contacts, facilitate smooth installation of stations, help lay the ground work for effective education and outreach efforts during and after installation, start educational efforts to introduce students and teachers to background Earth Science content, raise public awareness, and provide public information. E&O is a critical component of public relations and education at the site identification / permitting stage. The siting and permitting team should include several individuals, one of whom should have significant E&O expertise and can talk with landowners, schools etc and whose job it is to coordinate closely with other USArray E&O activities. This person can make initial contacts with many groups with whom the E&O program will need to work - schools, landowners, state surveys, forest service etc. During the pre-installation phase, the USArray web site will need to carry site specs and have a mechanism by which individuals and organizations can contact facility operators if they are interested in hosting a site. Early discussions with state geological surveys, USGS, CNSS, forest service, local, state, and national parks, local colleges and museums can maximize coordination of USArray E&O efforts with existing programs run through these organizations. Early identification of one or more museums for displays to be in place during the experiment in that region is important so that exhibits can be under development before the array "comes to town". Where possible, museums might also be a site for a station. The USArray E&O team will need to work with local teachers and scientists to develop educational and media materials specific to each new region. Web site development and maintenance will be an important part of these activities. The E&O team will also assist in training of participating scientists in public relations, providing them with information on E&O efforts and opportunities so that they can communicate these effectively.

    E&O Pre-Installation Activities could include:

    6.4 Education and outreach efforts during the deployment (while array is operational) can be used to, increase profile of Earth sciences, educate the public as to how science is done, and involve educators and students in hands-on science. These activities will require, close coordination with the IRIS DMC that will maintain data archives and with array operations to provide real-time data feeds to schools, museums, and the media. E&O personnel will coordinate efforts to identify earthquakes of interest, produce "teachable moments", and alert schools and the media to stations with high-quality data. They can also coordinate efforts to design educational materials as data start coming in, making tomographic images into usable educational materials, ensure web access to data for non-specialists, and produce earthquake reports for local and national media.

    E&O deployment/operations activities could include:

    6.5 Education and outreach activities subsequent to deployment of the array in given are important to provide follow-up for the community, students and educators, involve students and educators in analyses of data, and assimilate results from one region in preparation for deployment in the next region.

     

    E&O activities subsequent to installation of the array include:

    6.6 The Virtual Seismic Network Explorer. As an extension to the VSN under development for USArray, we propose to develop the VSN Explorer, a user-friendly, versatile interface to the VSN, along with accompanying educational materials for museums and the K-12 classroom. These two environments have been selected for several reasons. First, most of the software development for the VSN Explorer will be for "behind the scenes" data handling and delivery that is invisible to the end-user. The flexibility of our proposed software infrastructure will allow most of the software to be shared, so that relatively minor adjustments can be made in the "front-end" services, to meet the exact specifications desired by museums and K-12 teachers. Second, the IRIS E&O program has already devoted significant resources to the development of museum exhibits and to K-12 and has developed critical contacts within those communities to facilitate the development, testing and evaluation of the VSN Explorer and educational materials. Third, the combination of museum and K-12 classroom audiences will enable a far-reaching and lasting impact.

    For both the museum and K-12 classroom environments, the development and effective implementation of the VSN Explorer will involve software development to produce the VSN Explorer interface and development of accompanying educational resources (software, hardcopy and electronic materials). To ensure museum and K-12 products which meet the needs of the target audiences, an advisory group will be constituted comprising K-12 educators, science education specialists, museum exhibit experts and museum education experts, in addition to seismologists and members of the IRIS E&O program. We will use our existing contacts at the Franklin Institute, the American Museum of Natural History (AMNH), the New Mexico Museum of Natural History and Science (NMMNHS), and the Carnegie Museum in Pittsburgh, to establish a strong museum presence in the advisory group. We will draw on teachers with whom we have worked through IRIS E&O workshops and through our contacts in the Princeton Earth Physics Program (PEPP, a program to place seismometers in high schools across the U.S.) for K-12 teaching and classroom expertise. The advisory group will provide input and feedback at all stages of development, testing and evaluation of the VSN Explorer and associated educational materials.

    The VSN Explorer will require a PC (Windows or Macintosh machine) with a WWW browser and Internet access. It will provide a versatile, web-based interface to the VSN that delivers real-time seismic data from all over the world in a variety of displays. The VSN Explorer is envisioned to have the following capabilities:

     

    For all of these capabilities and products, there will be substantial user control and interactivity so that the user (teacher, student, museum leader or visitor, interested person on the Internet) can explore and investigate earthquakes, seismology, plate tectonics, related Earth science topics using the VSN and the VSN Explorer.

    The VSN Explorer in Museums. The VSN will provide opportunities for significant extension of the capabilities of our current museum displays through enhanced access to real-time seismic data from around the world. Currently, near-real-time data displayed on the triple drum recorders in our exhibits is available from only a limited set of seismic stations and data from only three stations can be displayed simultaneously. The Seismic Monitor, a computer based map display of global seismicity updated every few minutes, lacks flexibility in its map displays. The development of the VSN Explorer would allow museums to tailor the computer and drum-recorder displays to their local area, or to retain a more global focus, or to switch between a regional and global focus as desired. Real-time seismic data displayed on the drum recorders could be supplemented by real-time data from additional user-selected stations displayed on a computer monitor.

    Recently, we have developed the capability for "playback" of the Northridge earthquake on the drum recorders of our museum display. Typically, we set the three drum recorders to display seismograms for the Northridge earthquake recorded at three seismic stations across the U. S.. Several important concepts can be demonstrated using this playback, such as the travel-time of seismic waves across the U.S., the characteristics of different types of seismic waves, the principles used in locating an earthquake, and the characteristics of aftershocks. The VSN Explorer would extend the playback capability to include ANY recent or historical major earthquake and would extend the display capability (either on the drums or on a computer monitor) to any choice of seismic stations.

    Prototypes of the museum display interface will be incorporated into our existing displays at the Carnegie Museum and the AMNH for testing and evaluation. We will work closely with the education groups at these museums to develop web-based and hard-copy supplementary educational materials to be used by museum’s educational programs in conjunction with the IRIS exhibit. Both the Carnegie Museum and the AMNH already have well-established evaluation procedures, and we will use their expertise in helping us design effective formative and summative evaluations. Because we support our existing museum displays on an ongoing basis, we have already established a precedent for incorporating feedback from formative evaluations into updates and changes to both existing and new displays.

    The VSN Explorer in the K-12 Classroom. The VSN will extend and enhance current IRIS E&O products and activities for the K-12 classroom through the development of the VSN Explorer. The versatile, interactive nature of the VSN Explorer will enable K-12 students and teachers to have access to continuous monitoring of local, regional and global seismicity bringing the excitement of science and discovery into the classroom whenever a local or large regional or global event occurs. Important components of the VSN Explorer for the K-12 classroom will be easy selection of seismic stations comprising the desired virtual network, the real-time map and seismogram display features, an ability to easily replay recent archived seismic events such as a large earthquake recorded the previous day, and an easy interface between the real-time displays and educational software tools that use the seismic data.

    As with the museum exhibit display, the exact form of the end product for the K-12 classroom will be decided by the advisory group. Input from teachers in this group will be critical and these teachers will receive prototypes of the VSN Explorer to test in their schools and to enable feedback during the developmental period. However, simply placing VSN Explorer in K-12 classrooms will not ensure effective education. For the VSN Explorer to be useful in this setting, auxiliary teaching materials including software tools and supplementary hard copy materials must be developed incorporating Earth science content and effect pedagogical approaches. For example, software tools useful for manipulating seismic data will need to be adapted to be age-appropriate. IRIS E&O has already developed materials for K-12 education covering topics in seismology and related Earth science. Through collaboration with the advisory group these materials will be adapted and supplemented to form a package suitable for use with the VSN Explorer. Teachers in the advisory group will test these materials in their classrooms along with the VSN Explorer prototypes and software tools. Formative evaluation will include questionnaires, follow-up meetings of the full development group and site visits to schools represented in the advisory group. Software and hard copy materials will be modified accordingly.

    Following pilot testing by the advisory group, the VSN Explorer and accompanying materials will be field tested by an independent group of 20-25 teachers. Each teacher will be provided with a package composed of the VSN Explorer and software tools installed on a PC, the accompanying hard copy materials, questionnaires and evaluation forms. The VSN Explorer and materials will be demonstrated in a workshop. Follow-up will include site visits and a summative evaluation of the field test will be performed by an external evaluator. At the end of this development program the VSN Explorer and associated software and educational materials will be ready for broad dissemination in the K-12 and post-secondary communities.

    Appendix 7: USArray and Measurements on the U.S. Continental Margins

     

    John A. Orcutt

    Director, Cecil H. and Ida M. Green Institute of Geophysics and Planetary Physics

    Scripps Institution of Oceanography

    La Jolla, CA 92093

    Jorcutt@igpp.ucsd.edu

     

    The proposed permanent and mobile stations in the USArray program present an extraordinary opportunity for understanding the lithosphere, asthenosphere, and upper mantle structure and dynamics beneath North America. The ties to societal concerns about earthquake and volcanic hazards are obvious. However, the continents and the dynamic processes responsible for the construction and evolution of the North American continent do not stop at the water's edge. Throughout geological history, active continental margins have been the sites at which continents have themselves been generated, although details of the processes by which this occurs remain enigmatic. They have also been the zones in which the largest earthquakes occur - a modern analog is the Pacific Northwest; to oceanographer, the northeastern Pacific. It is essential that an integrated program be mounted to collect the necessary data not only on the continents, but also on the surrounding submerged continental margins and seafloor. The tools now exist for such studies and the National Science Foundation has established a national pool of instruments that should be used to support USArray.

     

    The oceanographic community and the NSF elected to establish a pool of seafloor seismic instruments for a variety of reasons. Advances in microprocessor and storage technology have made a new generation of seafloor instruments possible. The seismometers and hydrophones used in the instruments are now much broader-band (10 MHz - 40 Hz) and the instruments can be deployed unattended for more than a year. Many of the scientific techniques developed by the US seismological community to take advantage of the fixed and portable seismographs in the IRIS GSN and PASSCAL programs are in the process of being transported to the oceans. The rationale for the national pool was developed at a Workshop at the Monterey Bay Aquarium Research Institute (MBARI) in July 1997 ( http://victory.ucsd.edu/OBS_Meeting_Site/OBS_Report.html ). Following the publication of this report, the NSF requested proposals from academic institutions to provide instruments to a national pool. The Scripps Institution of Oceanography and the Woods Hole Oceanographic Institution were selected as Institutional Instrument Contributors (IIC) and the Lamont Doherty Earth Observatory may join the pool at a later date.

     

    The current pool consists of 150 OBS/H with hydrophones and vertical geophones for active source studies and 50 OBS with broadband sensors (see above) which can be deployed for more than a year at a time. Of course, the 50 OBS can be added to the 150 OBS/H to support an active experiment with 200 instruments. USArray could make very fruitful use of the 50 four component instruments for long-term deployments in conjunction with the continental, portable array which will be moved around the continental US during the coming decade. These instruments are all 24-bit, micropower seismographs with precision clocks with a drift rate of no more than 0.1s/Yr.

     

    The instruments are freely available to anyone with NSF funding - the pool operators will provide the instruments requested as well as the manpower to launch and recover. The data will be subjected to quality control at Data Collection Centers at the two IIC's. The IIC's will provide the data to the IRIS DMC for distribution. The PI's can place a proprietary hold on the data for up to two years, but for the USArray, the data should be immediately available for the seismological community. The cost of deploying 25 of the four component instruments for a full year is $391,153 for budget planning purposes. The IIC's are supported by Cooperative Agreements with the NSF and the actual instrument costs will not be included in proposals although the PI is required to include a cost estimate (provided by the national pool) in the proposal.

     

    More information can be found at http://victory.ucsd.edu/obsip.html. Bob Detrick at the Woods Hole Oceanographic Institution is the first Chair of the national pool Management Committee. The potential role of permanent stations on the continental margins can be explored at http://victory.ucsd.edu/whitepaper/front.html .

     

     

    Appendix 8: Synergy with other EarthScope Initiatives.

    NOT FINISHED - PLACE HOLDER

    Subsequent to the first USArray workshop, Program Officers of the Division of Earth Sciences (EAR) united several intertwining facilities initiatives into a single integrated effort known as "EarthScope - A Look into Our Continent". EarthScope includes USArray and initiatives for a Plate Boundary Observatory, the San Andreas Fault Observatory at Depth (SAFOD); and Interferometric Synthetic Aperture Radar (InSAR). To first order USArray is a continental scale seismic array to provide coherent 3-D image of lithosphere and deeper Earth. The seismic array provides infrastructure that is easily leveraged serving as a natural platform for additional instrumentation and multidisciplinary investigations across the continent. SAFOD is a borehole observatory across the San Andreas Fault to directly measure the physical conditions under which earthquakes occur. PBO is a fixed array of strainmeters and GPS receivers to measure real-time deformation on a plate boundary scale and InSAR provides spatially continuous strain measurements over tectonically active regions. Collectively, EarthScope is a distributed, multi-purpose set of instruments that significantly expands our observational capabilities, providing synoptic images of the solid Earth. It represents an extremely exciting new set of tools for research in earthquake physics, volcanic processes, core-mantle interactions, active deformation and tectonics, continental structure and evolution, geodynamics, and fluids (magmatic, hydrothermal, and meteoric) in the crust.

    USArray and SAFOD

    The USArray and SAFOD projects have fundamental goals in common. Both aim to provide information on the structure and rheology of the Earth in three dimensions over a range of spatial scales. Both will investigate the processes and conditions that control deformation of the Earth. SAFOD will provide critical "in-situ" data on physical and rheological properties of earth materials at depth. USArray provides the tool for extrapolating the in-situ data to a broader region and larger spatial scales. A substantial amount of information exists about the crustal volume containing the fault at Parkfield, but much remains to be learned if we are going to be able to build a comprehensive structural and physical model of the fault zone. USArray allows for the multiple-scale, multi-disciplinary surface observations that both put the SAFOD observations into a broader geoscientific context and allow an extrapolation of the in-situ properties derived by SAFOD on one part of the San Andreas fault to other parts of the fault, and ultimately to the crust and fault zones elsewhere.

    USArray and PBO (need text)

    Fixed grid transportable array provides link to lithospheric and deeper mantle structure

    Flexible array instruments provide substantial arsenal for focused high-resolution studies of fault zones and crustal structure

    Infrastructure can be leveraged for additional GPS measurements

    Appendix 9: Viewgraphs/Overheads