Grand Challenges for Seismology

Document Sample
Grand Challenges for Seismology Powered By Docstoc
					                                                   Eos, Vol. 90, No. 41, 13 October 2009

                                                                                                       VOLUME 90            NUMBER 41
                                                                                                       13 OCTOBER 2009
EOS, TRANSACTIONS, AMERICAN GEOPHYSICAL UNION                                                          PAGES 361–372

Grand Challenges for Seismology                                                                       in slow events that occur surprisingly regu-
                                                                                                      larly, accompanied by low-amplitude seis-
                                                                                                      mic tremor (Figure 1). There are many spe-
PAGES 361–362                                       environment, discovering and mapping nat-         cific issues remaining to be addressed, such
                                                    ural resources, contributing to national and      as the relationship of this episodic slip and
   Seismology is the study of the propaga-          international security, and understanding         tremor to major earthquakes, and achiev-
tion of elastic waves, the sources that gener-      the dynamic processes in the interior of the      ing a detailed physical understanding of the
ate them, and the structures through which          Earth. This article provides a summary of         nonlinear processes by which faults slip is a
they propagate. It also is a fundamental, high-     the 10 grand challenges.                          major challenge.
resolution tool for exploring the interior of                                                            How does the near- surface environment
the Earth from crust to core, as well as other      Specific Challenges                               affect natural hazards and resources? The
bodies in the solar system. A remarkable                                                              location and severity of many natural haz-
diversity of multidisciplinary societal applica-       How do faults slip? The steady motions         ards are strongly influenced by near- surface
tions of seismology has emerged, including          of the tectonic plates build up stresses that     material properties. Determining Earth’s his-
hydrocarbon and resource exploration, earth-        are relieved mainly through slip on faults.       tory of natural climate change relies in part
quake detection and hazard assessment,              Recent observations have revealed a rich          on seismic imaging of shallow sedimen-
nuclear test monitoring and treaty verifica-         spectrum of fault behavior, ranging from          tary deposits that record and respond to
tion, volcano and tsunami warning systems,          steady sliding with little apparent resis-        climate variations. Near- surface processes
and aquifer characterization. New directions        tance to earthquakes that can slide at super-     affect water, energy, and mineral resources
in seismology are evolving that are relevant        shear velocities (faster than the speed of        at depths of meters to a few kilometers.
to climate and environmental change, such           S waves in the rocks) and that can emit           Detailed knowledge of the Earth’s near sur-
as resolving fine- scale seismic stratigraphy,       shock waves that may cause exceptionally          face is therefore a crucial part of managing a
monitoring carbon sequestration, detect-            damaging ground motions. Only in the past         sustainable environment for civilization. One
ing sudden movements of glaciers and ice            decade has it been discovered that major          of the most important challenges for seis-
sheets, mapping the internal fine structure of       parts of some fault systems slip repeatedly       mology is to understand how strong ground
the ocean, and reconstructing the twentieth-
century history of global storm activity from
ocean-generated seismic noise.
   The broad scope of seismological
research positions the discipline to contrib-
ute significantly to the U.S. National Science
Foundation (NSF) Directorate for Geosci-
ences’ emphases on dynamic Earth pro-
cesses and climate change in the 2010 bud-
get request to Congress.
   The seismological community, through a
community workshop and writing commit-
tee process sponsored by NSF, earlier this
year identified 10 grand challenges for seis-
mology at the forefront of research on Earth
systems. The resulting document, “Seismo-
logical grand challenges in understanding
Earth’s dynamic systems,” was published in
early 2009 and is available online at http:// The report is directed
at a broad readership, including researchers
and educators in other disciplines as well as
a general academic and government audi-
ence. The report includes a number of side-
bars illustrating recent discoveries and appli-
cations that are appropriate for classroom
and other broad use.                                Fig. 1. Location of migrating tremor during a 2- to 3-week episode of slow slip on the Cascadia
   The grand challenges, which are framed           subduction zone. Most of the relative plate motion in the slow slip area is accommodated by
by fundamental research issues, encompass           similar slip events that repeat approximately every 14 months. Plate boundary slip in the “locked
mitigating natural hazards, monitoring the          zone” to the west of the contours of partial locking occurs during great earthquakes such as
                                                    the Mw ~9 Cascadia megathrust earthquake in 1700. Locking refers to the percentage of slip
BY D. W. FORSYTH, T. LAY, R. C. ASTER,              between plates that occurs in stick-slip events as opposed to gradual, nearly continuous creep.
AND B. ROMANOWICZ                                   Image courtesy of A. Wech and K. Creager.
                                                  Eos, Vol. 90, No. 41, 13 October 2009
motions are produced by earthquakes                and how much water is stored in the mantle      core? The thermal evolution of the Earth, the
and to translate this understanding into           (which may amount to more than five ocean        driving forces of plate tectonics, and the gen-
improved hazard maps. Nonlinear responses          volumes) and whether changes in mineral-        eration of the magnetic field all involve con-
to shaking, such as soil liquefaction, and the     ogical phase lead to greater concentrations     vective flow in the mantle and core. Improv-
complex pattern of strong ground motions           of water in the mantle transition zone and      ing the seismological resolution of deep
can be predicted with comprehensive three-         cause regions of partial melt near the global   structure as data accumulate and as new
dimensional (3- D) modeling of potential           410-kilometer- deep discontinuity.              analysis methods are developed will help
earthquakes and knowledge of soil proper-             How do magmas ascend and erupt? Seis-        reveal the patterns of flow. Recent observa-
ties, an undertaking that straddles the inter-     mological monitoring is one of the primary      tional studies, combined with mineral phys-
face between seismology and earthquake             ways of forecasting or predicting volcanic      ics experiment and theory, have shown that
engineering.                                       eruptions. An increase in microearthquake       large- scale chemical heterogeneity is pres-
   What is the relationship between stress         activity and harmonic tremor, or changes in     ent in the mantle and that the interaction of
and strain in the lithosphere? Plate tecton-       seismic velocity as moving magma changes        compositional and thermal buoyancy must
ics provides the kinematic framework for           the shape of the volcano and fractures the      be considered in modeling convective pro-
describing rates of deformation, but it does       surrounding rock, often precedes eruptions      cesses. The large- scale 3-D elastic struc-
not quantitatively account for how plates          by several days, providing some warning         ture of the mantle is now fairly well known,
move and deform. Rheology describes the            of an eruption. Current eruption prediction     but where detailed studies provide higher
linkage between the forces (stresses) and          methods are primarily empirically based,        resolution, pronounced sharp or short-
the resulting deformation (strains). Motions       however, because magma plumbing systems         wavelength features are found. This suggests
and strains now are precisely measured with        are poorly known. A major challenge is to       that small- scale convection plays a critical
satellite imaging and networks of Global           improve scientific understanding and predic-     role in the dynamics of Earth’s deep interior.
Positioning System receivers, strainmeters,        tion capabilities through better determina-        How are Earth’s internal boundaries
seismometers, and tiltmeters, but the caus-        tion of the physical changes that accompany     affected by dynamics? Internal boundaries
ative stresses only can be inferred. Meeting       eruptions, including improved imaging of        in Earth (and other planets) are associated
the grand challenge of understanding the           the interior of volcanic systems and quanti-    with the primary compositional layering that
stress distribution and the temporally and         tative characterization of magma migration      resulted from the chemical differentiation
spatially dependent rheology is necessary          and eruption processes.                         of the planet and with mineralogical phase
to unraveling how some earthquakes trigger            What is the lithosphere- asthenosphere       changes controlled by pressure and tem-
other earthquakes thousands of kilometers          boundary? The lithosphere is Earth’s            perature variations. These boundaries may
away or how, for instance, the great Sumatra       mechanically strong outer shell that makes      be deflected by convective processes, thus
earthquakes of 26 December 2004 (seismic           up the tectonic plates, underlain by the        providing clues to the location and inten-
moment magnitude Mw = 9.3) and 28 March            weak asthenosphere, which flows and              sity of upwelling and downwelling. Because
2005 (Mw = 8.7) were coupled.                      deforms to accommodate plate motions. The       changes in rheology, composition, and den-
   How do processes in the ocean and atmo-         lithosphere often is thought of as the ther-    sity occur across the boundaries, they can in
sphere interact with the solid Earth? Ocean        mal boundary layer between the cold sur-        turn exert a strong influence on the pattern
storms, bolides, tornadoes, and glacier            face of Earth and the planet’s hot interior,    of convection. The challenge for seismology
calving all generate signals that are readily      but recent studies have shown that there is     is to map these boundaries, including their
detected by seismometers and atmospheric           often a sharp seismic discontinuity at the      3- D topography and sharpness, which are
infrasound recorders. The multidisciplinary        base of the lithosphere inconsistent with a     key clues to quantifying their mineralogical,
topic of how processes in the ocean and            simple gradual thermal transition. Changes      thermal, and compositional nature and to
atmosphere couple into seismic waves and           in composition, volatile content, and anisot-   interpreting the dynamic processes that con-
how these waves can be used to monitor             ropy of the mantle, and perhaps the pres-       trol these variations.
the global environment is one of the high-         ence of melt, may play roles in creating the
priority challenges. Recently, it was estab-       discontinuity. Lithosphere- scale seismol-      Requirements to Meet the Challenges
lished that the Earth’s long-period “hum” of       ogy is being revolutionized by new data
free oscillations continuously excited at peri-    from large- scale seismometer deployments,         The report “Seismological grand chal-
ods of hundreds of seconds is generated by         such as the USArray component of the NSF-       lenges in understanding Earth’s dynamic
midlatitude winter storms through an as yet        funded EarthScope project, and by new           systems” describes the detailed seismologi-
poorly understood mechanism. On the other          analysis techniques. However, many chal-        cal approaches and practical requirements
end of the seismic frequency scale, active         lenges to understanding the evolution and       needed to make progress in attacking each
sources used in seismic profiling, such as          structure of the lithosphere and the astheno-   of the 10 grand challenges. A number of
in routine imaging of subseafloor structure,        sphere remain.                                  common themes emerge in terms of these
can detect layering and mixing in the water           How do plate boundary systems evolve?        approaches and needs. For example, increas-
column itself. The images’ unprecedented           Most earthquakes and volcanoes occur at         ingly massive data sets, inversions for 3-D and
horizontal resolution can help with under-         plate boundaries. Most of the deformation       4-D multiscale models, and realistic simula-
standing internal waves, turbulent mixing,         and volcanic activity at plate boundaries in    tions incorporating as much of the physics
and ocean circulation.                             the oceans may take place in a zone only a      as possible require enormous computational
   Where are water and hydrocarbons hid-           few hundred meters across at the surface,       capabilities. Thus, collaborative efforts to
den beneath the surface? Seismological tech-       yet plate boundary systems may be hun-          increase access to state- of-the-art computing
niques have long been used to map aqui-            dreds of kilometers wide in the continents.     are essential. Another such theme is seismol-
fers and explore for hydrocarbon resources.        The geometry of these diffuse boundaries        ogy’s long tradition of open access to all data
Modern exploration seismology methodolo-           changes with time, and the areas within         sets and storage of these data sets in perpe-
gies, including 4-D (time lapse) mapping,          the boundary system that are most active        tuity; this approach needs to be encouraged
routinely are used to monitor the extraction       may shift. Coordinated seismological, geo-      and supported globally. Networks of perma-
and movement of hydrocarbons in real time          detic, geomorphological, deep drilling, and     nent, broadband, real-time observatories
on land and at sea. Similar approaches now         geological studies are needed to meet the       form a backbone of national and worldwide
are being applied to investigate the poten-        challenge of determining what controls the      monitoring efforts, and these networks need
tial for carbon dioxide sequestration, and         location, width, and activity of dynamically    to be maintained, upgraded, and, where pos-
these approaches will be critical for manag-       evolving plate boundaries.                      sible, expanded to the oceans.
ing these efforts. Looking deeper, there is           How do thermal and compositional varia-         An essential need for several of the
great interest at present in deducing where        tions control convection in the mantle and      grand challenges is the availability of large
                                                   Eos, Vol. 90, No. 41, 13 October 2009
pools of portable instruments for seismo-           the academic community. Further, the new          seismology and other disciplines need to
logical investigations of continental and           seagoing R/V Marcus G. Langseth (owned            be fostered and strengthened. Progress on
oceanic environments at higher resolution           by NSF and operated by Lamont- Doherty            the seismological grand challenges noted
than that afforded by the current global            Earth Observatory of Columbia University)         here, and on the many societal applica-
network of permanent stations. The pools            needs to be fully supported in a way that         tions of seismology, hinges on improved
of three- component, short- period, and             makes its 3-D imaging capabilities more           interdisciplinary interactions and commu-
broadband sensors need to be expanded,              readily accessible to investigators.              nications, in addition to the shared, practi-
in the oceans and on land, for the next               There is a common need for the devel-           cal requirements described above.
generation of 3-D and 4-D imaging efforts           opment and coordination of advanced
                                                                                                        —DONALD W. FORSYTH, Brown University, Provi-
of crustal, lithospheric, and deep mantle           data products to make the results of seis-        dence, R. I.; E-mail:;
and core structure.                                 mological research more accessible to             THORNE LAY, University of California, Santa Cruz;
   In addition, a new facility should be            the public and to Earth scientists in other       RICHARD C. ASTER, New Mexico Institute of Mining
established to make controlled seismic              disciplines. Finally, strong synergisms           and Technology, Socorro; and BARBARA ROMANO-
sources for land studies more available to          within the Earth science arena between            WICZ, University of California, Berkeley

                                                                                                      opportunity to use the full range of seismic
Probing the Hawaiian Hot Spot With New                                                                techniques that have been applied success-
Broadband Ocean Bottom Instruments                                                                    fully in land- deployed experiments. Body
                                                                                                      wave and surface wave tomographic imag-
PAGES 362–363                                       a multidisciplinary program whose center-         ing as well as receiver function and compli-
                                                    piece is a large network of four- component       ance analyses will provide new constraints
   The Hawaiian hot spot is regarded as the         broadband ocean bottom seismometers               on elastic and anelastic seismic structure
textbook example of the product of a deep-          (OBSs) and three- component portable              and major discontinuities from crustal
rooted mantle plume [Wilson, 1963; Morgan,          broadband land stations (Figure 1). Occupy-       depths into the lower mantle. The analysis
1971]. Its isolated location, far from any plate    ing a total of 82 sites and having an overall     of shear wave splitting and surface wave azi-
boundary, should provide an opportunity             aperture of more than 1000 kilometers, this       muthal anisotropy will help reveal mantle
to test most basic hypotheses on the nature         experiment is one of the first large- scale,       fabric and flow patterns.
of plume-plate interaction and related mag-         long- term deployments of the new broad-             Now, about 18 months after the last
matism [e.g., Ribe and Christensen, 1999].          band OBSs in the U.S. National Science            OBSs were recovered from the ocean
Yet the lack of crucial geophysical data has        Foundation–supported national OBS Instru-         floor, the high overall return and quality
sustained a debate about whether Hawaii’s           ment Pool (OBSIP). PLUME is providing an          of PLUME data allow for the production of
volcanism is plume-related or is instead the
consequence of more shallow processes,
such as the progressive fracturing of the
plate in response to extensional stresses
[Turcotte and Oxburgh, 1973].
   In the plume model for Hawaii’s volca-
nism, hot material is expected to ascend
near vertically within the more viscous sur-
rounding mantle before ponding and spread-
ing laterally beneath the rigid lithosphere.
Mantle convection in general, and the fast
moving Pacific plate in particular, shear
and tilt the rising plume. The plume top
is dragged downstream by the plate, and
this dragged material may give rise to an
elongated bathymetric swell [Davies, 1988;
Olson, 1990; Sleep, 1990; Phipps Morgan
et al., 1995]. However, identifying the domi-
nant cause of the swell remains elusive, and
proposed mechanisms include thermal reju-
venation, dynamic support, compositional
buoyancy, and mechanical erosion (see Li
et al. [2004] for a summary). There is also
considerable debate about the continuity of
the plume within the mantle, how discrete
islands are formed, and how a deep-rooted
plume interacts with the mantle transition          Fig. 1. Site locations of the two deployment phases of the Hawaiian Plume-Lithosphere Under-
zone [e.g., van Keken and Gable, 1995].             sea Mantle Experiment (PLUME). Also shown are sites of permanent stations of global seismic
                                                    networks relevant to this study. Station KIP (Kipapa, Oahu) is jointly operated by the French Geo-
                                                    scope program and the U.S. Geological Survey (USGS); POHA (Pohakuloa, Hawaii) is operated
Seismic Imaging of Hawaiian Mantle                  by USGS; and MAUI is operated by the German Geo-ForschungsNetz (GEOFON). USGS station
                                                    MIDW (Midway; see Figure S1 in the electronic supplement) is not shown. Phase 1 operated from
   Seismic imaging can help distinguish             January 2005 through January 2006, and phase 2 operated from April 2006 through June 2007.
among plausible models, but the deploy-             Two sites with unrecovered ocean bottom seismometers (OBSs) from phase 1 and six sites from
ment of seismic stations that has been lim-         phase 2 were visited by Woods Hole Oceanographic Institution’s remotely operated vehicle (ROV)
ited to the nearly aligned Hawaiian Islands         Jason in November 2007.The OBSs at sites 57 and 59 were recovered at that time. Four sites
has so far led to incomplete images of the          with five lost OBSs (sites 9; 42, with two OBSs; 52; and 72) remain unvisited. Open numbered
crust and mantle beneath and around                 circles mark instruments with a loss of differential pressure gauge (“no DPG”) and/or vertical-
Hawaii. The Hawaiian Plume- Lithosphere             component seismometer data (“no Z”). During the first deployment cruise, 11 dredge hauls were
Undersea Mantle Experiment (PLUME) is               performed at six locations to retrieve fossil corals and deep-rift volcanic rocks.

Shared By: