T54C-01
A Critical Stress State in Stable Continental Regions?
Despite considerable study we still do not understand why stable continental region (SCR) earthquakes occur where they do. Although most earthquakes can be related to structural features such as failed rifts, SCR regions contain many such features, few of which are active. Previous studies have suggested that SCRs are characterized by a critical stress state, such that stress in the upper crust is pervasively close to the failure stress of faults. We present a simple rheological model incorporating stress-dependent aseismic deformation that can explain how a critical stress state can be maintained. In a low strain-rate environment, even a low rate of aseismic deformation can significantly reduce the accumulation of stress available to drive earthquakes. Given a low strain-rate and a high failure stress, we show that it is possible to accommodate all strain by permanent deformation. However, one is left with the apparent paradox that, while borehole measurements suggest that intraplate crust is everywhere critically stressed, earthquakes (and triggered earthquakes) appear to occur preferentially in certain regions. The question is, what, if anything, is unusual about these regions? Although some lines of evidence point to local stress concentrations and/or to locally weak faults, we note that the effects of stress interactions are expected to affect seismicity rates for longer periods than in SCRs, and may be a first- order control on observed earthquake distributions during a short historical record. Thus, while there is a relatively compelling association between failed rifts and large earthquakes, the basis of differentiation of hazard within or between presently active structures and those that appear inactive is likely to be less compelling than the historical record inclines us to believe. http://pasadena.wr.usgs.gov/office/hough/AGU2007
T54C-02 INVITED
Deep Mantle Dynamics under the North American Continent Drives Localised Flow and Stress Below the New Madrid Seismic Zone
The origin of intraplate earthquakes represents one of the outstanding problems in modern geophysical research, and the major earthquake sequence that struck the central Mississippi River Valley in 1811-1812, the so-called New Madrid seismic sequence, has become a principal target of this research. As Johnston and Schweig (1996) have noted, the occurrence of such large magnitude earthquakes in "stable" North American crust, far from any plate boundaries, remains an enigma. To understand the possible origin of this enigmatic seismic activity we have developed a new high resolution model of mantle flow below North America. The model is constrained by simultaneously inverting global seismic and mantle-convection data sets and it includes an explicit treatment of the positive chemical buoyancy of the continental tectosphere. Moreover, it adopts a depth dependent mantle viscosity structure which reconciles both glacial isostatic adjustment (GIA) and convection data. The flow model successfully reproduces plate velocities and observations of surface gravity and topography, including the continent-scale quasi-linear depression (after corrections for GIA and crustal heterogeneity) extending from northern Alaska to Venezuela. The predictions also match lithospheric flow and stress fields inferred from local and regional measurements of seismic anisotropy and surface deformation. We demonstrate that these signals are largely driven by viscous flow coupled to density anomalies within the lower mantle associated with the descent of the ancient Kula-Farallon plate system. More importantly, the flow calculations elucidate how these large-scale heterogeneities give rise to flow and stress patterns below the New Madrid Seismic Zone which are favourably oriented with respect the local fault geometry in this portion of the Mississippi valley.
T54C-03 INVITED
Intraplate Seismicity Within "Stable" North America
The origin of earthquakes within stable continental regions (SCR"s) has been the subject of debate over the past thirty years. Here we examine the correlation of North American SCR earthquakes with four geologic and geophysical data. We find that: (1) Archean age (3.8-2.5 Ga.) North American crust is essentially aseismic, whereas post-Archean (less than 2.5 Ga.) crust shows no clear correlation of crustal age and earthquake frequency or moment release; (2) seismicity is correlated with continental paleo-rifts, and (3) seismicity is correlated with the NE-SW structural grain of the crust of eastern North America, which in turn reflects the opening and closing of the proto- and modern Atlantic oceans. Seismicity levels are very low to the west of the Grenville Front (i.e., in the Archean Superior craton). The correlation of seismicity with NE-SW oriented lineaments implies that some SCR seismicity is related to the accretionary and rifting processes that have formed the North American continental crust during the past 2 Gy. We further evaluate this hypothesis by correlating SCR seismicity with recently-obtained deep seismic reflection images of the Appalachian and Grenville crust of southern Canada. These images show numerous faults that penetrate deeply (40 km) into the crust. We conclude that the deep structure of the crust, in particular the existence of deeply penetrating faults, is the controlling parameter, rather than lateral variations in temperature, rheology, or high pore pressure. The distribution of SCR earthquakes in eastern North America is consistent with the existence of deeply penetrating crustal faults that have been reactivated in the present stress field. This implies that seismic hazard is more widespread in central and eastern North America than indicated by the limited known historical distribution of seismicity.
T54C-04
Intensity, magnitude, location and attenuation in India for felt earthquakes since 1764
We have subjected more than 7444 intensity reports from 204 earthquakes in India that have occurred since 1764 to a felt-intensity vs attenuation analysis using the methods of Bakun and Wentworth (1997). Our catalog consists entirely of re-evaluated MSK intensities from authentic accounts. We include the 43 north Indian earthquakes analyzed by Ambraseys and Douglas (2004) supplemented by an additional 201 earthquakes newly evaluated by the Amateur Seismic Centre, India. We compare first the approximately 90 earthquakes for which we have both an instrumentally determined location and moment magnitude, and examine attenuation effects throughout India. We also compare the intensity-derived epicenter with the instrumental location. From empirical regressions we calculate the most probable locations and magnitudes for earthquakes in the remainder of the catalog. For those earthquakes that occur on extended sources we have modified our search procedure to seek the most probable center of a line source representing the closest edge of a subsurface dislocation to the surface. During our talk we shall present the most interesting of our findings, with comparisons between Indian and North American attenuation laws, and associate our newly derived locations with physical processes within the Indian plate and along its boundaries.
T54C-05
Spatio-temporal complexity of continental intraplate seismicity: insights from geodynamic modeling and implications for seismic hazard estimation
Continental intraplate seismicity, which cannot be readily explained by plate tectonic theory, has remained puzzling. Observations from different continents show complex spatiotemporal patterns of intraplate seismicity: both spatial clustering in seismic zones and scattering across entire continents; temporal clustering followed by long period of quiescence; and migration of seismicity from one seismic zone (or region) to another. Thus assessment of earthquake hazard based on the limited historic record may be biased toward overestimating the risks in regions of recent large earthquakes and underestimating the risks where seismicity has been quiescent. We have simulated the spatiotemporal evolution of intraplate seismicity in a 3D viscoelastic finite element model. The model incorporates tectonic loading, crustal failure, and coseismic and postseismic stress evolution. With pre-specified perturbations of crustal strength, the model predicts various spatiotemporal patterns of intraplate seismicity: spatial clustering (in narrow belts) and scattering ( across large regions) in hundreds of years, networked seismic belts in thousands of years, and apparently randomly scattered events over tens of thousands of years. The model results also show temporal clustering of intraplate earthquakes in seismic zones. The predicted spatial-temporal clustering results mainly from coseismic and postseismic stress migration. By considering continuous strength weakening following a large earthquake, the model predicts temporal earthquake clustering on an individual fault even in the absence of far-field loading. These results provide useful insights into the complex spatial-temporal pattern of intraplate seismicity, and call for caution when assessing intraplate earthquake hazard based on the short spatio-temporal record of instrumentally recorded seismicity.
T54C-06
Great Intraplate Earthquakes and Glaciation: Effects of Rheology and the Tectonic State of Stress on Glacially Induced Faulting in Scandinavia
At the end of the last glaciation, northern Scandinavia experienced a dozen or so very large earthquakes, reaching magnitude 8. These reverse faulting events ruptured the surface in throws of 10 to 15 m, leaving faults scarps (some more than 100 km in length) that are still visible today. Although the deglaciation process is widely accepted as the cause of the earthquakes, little is known about the mechanics of the process and why they only occurred in northern Scandinavia, but not southern Scandinavia nor northern North America, for example. In this study we use 2D finite element modeling to investigate the effects of rheology and the tectonic state of stress on faulting potential during a glaciation. We use both a simple parabolic ice sheet model and a northwest-southeast profile through a three-dimensional model of the entire Weichselian glaciation. The earth models are based on the concept of an elastic plate overriding a viscoelastic half-space. We study how the response of the elastic lithosphere changes as we introduce layering and an increasing elastic thickness, from the thin oceanic lithosphere off the coast of Norway to the very thick cratonic lithosphere under Finland. Using established earth models derived from different data sets (glacial isostatic rebound, seismology and gravity) we see surprisingly small differences in the stability of faults at depths down to approximately 20 km. The pre-existing tectonic state of stress is critical when assessing fault stability. We show how a reverse and a strike-slip state of stress, respectively, influence the time history of fault stability during a glacial cycle. As the current stress state has likely not changed markedly in the last 10,000 years, we compare the modeled stress states to current stress estimates from inversion of earthquake focal mechanisms and deep boreholes. Hydrology at the base of ice sheets is a complex topic, but because pore pressure is important in fault mechanics, we explore the effects of simple static end-member models of possible pore pressures in the shallow crust during glaciation on fault stability.
T54C-07
Rapid Changes in Sediment Transport and the Potential for Interplate Earthquake Generation: The Cases of the Mississippi and Modern Yangtze
Sediment catchment basins at the mouths of the world's largest rivers generally do not have substantial seismicity, as they occur at ‘passive' continental margins. The rates of sediment transport to these basins can be substantial: in the near shore zone, of order 1 meter per kyr; and in deltaic deposition centers, of order 10-20 meters per kyr. While such systems generally have faulting associated with them, these are generally manifested as shallow growth faults having relatively low levels of seismicity. However, when significant changes in the sediment rate exist, such as at the time of the post-glacial flooding associated with the termination of the Last Glacial Maximum in North America, or when significant anthropogenic intervention (large dams) cause sediment transport to drop to effectively to zero in one location and to larger values at another location, the perturbation to the failure stresses that generate large earthquakes is enough to drive large earthquakes. With a viscoelastic layered model that matches local GPS array data [Ivins, Dokka and Blom, 2007: 34, L16303, doi:10.1029/2007GL030003], we determine the Coulomb stresses in the crust that are generated by the late Pleistocene, Holocene and present-day Mississippi sediment load changes and show that the greatest earthquake potential occurs during rapid sea-level rise when delta depocenters migrate landward. For the Yangzte in China, where the Three Gorges Dam impedes transport to the mouth of the river, by roughly 0.5 cubic kilometers per year [Yang et al., 2005: J. Geophys. Res., 110, F03006, doi:10.1029/2004JF000271], the perturbation is about a half MPa, or the same magnitude as the stress perturbations that link large earthquakes in active interplate deformation regimes. We shall present predictions of time-dependent failure Coulomb stress evolution for the latitude 30° N and east west profile: 100 – 120° E, central to coastal China.
T54C-08 INVITED
GPS Contribution to New Madrid Seismic Zone Hazard Estimation
Continued study of the New Madrid seismic zone further deepens the enigma of the world's most active intraplate seismic zone. Paleoseismic studies indicate the occurrence, with an approximate 500 year recurrence interval, of three to four previous events or sequences of several clustered events. Newer estimations of the magnitudes of the events in the 1811-1812 sequence suggest they are smaller than initially proposed, which lessens the amount of energy that has to be stored in the crust to produce the earthquakes by at least two orders of magnitude. Application of GPS technology (both campaign and continuous GPS measurements) to the problem of understanding New Madrid seismic zone earthquakes has not yet produced a definitive determination of a strain field there. Challenges in the application of GPS in the New Madrid seismic zone range from the lack of a paradigm for intraplate earthquakes to the effects of the unconsolidated Mississippi Embayment sediments on the transferal of crustal deformation through the sediments at large scale to monument stability at small scale. Both more time and denser continuous GPS measurements are needed before strong results from GPS can significantly change the seismic hazard estimates obtained independently by non-GPS studies.