T31G-01 INVITED 08:00h
Transpressional Mechanics of West-Central California
Analyses of seismicity, seismic reflection and geodetic data, and map-scale patterns of active faulting in west-central California do not support the hypothesis that plate-boundary stress is regionally refracted across a weak San Andreas fault, resulting in "fault-normal compression" and partitioning of plate motion onto parallel strike-slip and thrust faults. Specifically, this model is not consistent with the following observations: (1) Dextral faults of the eastern San Andreas system (i.e., the Greenville, Mt. Lewis, Northern Calaveras, Concord and Green Valley faults) typically strike 10-20° more northerly than the San Andreas and Hayward faults, but despite this apparent "releasing" geometry do not accommodate components of extension; (2) Major late Cenozoic folds commonly are oriented about 45° to the strike-slip faults, rather than parallel to the faults as assumed by some partitioning models; (3) Anticlines typically exhibit right-stepping, en echelon patterns, typical of dextral wrench tectonics; and (4) Loci of crustal shortening primarily are located in restraining bends or restraining step-overs between major strike-slip faults. These relations are consistently expressed along a SE to NW traverse that follows progressive transfer of dextral slip among segments of the Tesla-Ortigalita, Greenville, Concord, Green Valley and Maacama strike-slip faults for a distance of about 200 km. Rather than discrete partitioning of strike-slip and thrust deformation predicted by the "fault-normal compression" model, these observations document a 10°-20° CW rotation of macroscopic dextral shear from west to east across the Pacific-Sierran plate boundary (Unruh and Lettis, 1998). Viable mechanical models for transpressional tectonics in west-central California must account for this observed rotation of the principal strain rates (and presumably principal stresses). Because velocity gradients can be measured across the plate boundary with high precision, a logical modeling approach would be to test whether a constant velocity boundary condition and variable viscous rheology could produce the observed variations in strain rate and strain geometry.
T31G-02 08:20h
Structural Control on the Seismotectonics of the Coast Range-Great Valley Boundary Zone
Structural relations inferred from analysis of three-dimensional velocity-hypocenter inversions and seismic reflection data at several locations between lat $36.5\deg$ N and $38.5\deg$ N demonstrate the strong influence of pre-existing crustal structure on the kinematics of active strike slip and thrust faults located in the Coast Range-Great Valley Boundary Zone. The dextral Ortigalita and Greenville faults, which strike about N30W, intersect thick Great Valley crust at their northern limits and step left to the Greenville and Concord-Green Valley strike-slip faults, respectively. These restraining left steps are expressed as the Mt. Oso and Mt. Diablo anticlines and associated blind thrust faults. Strike-slip seismicity occurs beneath the Sacramento-San Joaquin delta at depths of 15-25 km, but steps left across the Potrero Hills and Cannon Hills anticlines to the Cordelia and Wragg Canyon strike-slip faults in the northern Coast Ranges. Active thrust faults extend eastward from the base of these strike-slip faults as right-stepping Reidel shears along pre-existing thrust faults that were the likely sources of the 1892 Winters-Vacaville earthquake sequence. Slip rates on the thrust faults typically are 1/4 or less of the 2-3 mm/yr slip rates of the strike-slip faults. South of $37\deg$ N, however, the Laguna Seca blind thrust has a slip rate of 2 mm/yr while the adjacent southern Ortigalita fault has a slip rate of 0.5 mm/yr or less. Near $37\deg$ N, the dextral Ortigalita fault makes a right step across a $\sim$5-km-wide pull-apart basin that is adjacent to two blind thrust faults beneath the monocline on the east flank of the Diablo Range. The apparently peculiar position of this pull-apart basin adjacent to active thrust faults is explained by systematic variations in crustal strength that appear to control the position and evolution of the Ortigalita fault. The Ortigalita fault is required to step right to work around a locally stiff (high-velocity) body beneath the pull-apart basin. This high-velocity ($\sim$6 km/s) body, possibly a gabbro intrusion responsible for the local outcrop of basalt (Basalt Hill), separates the northern and southern Ortigalita fault segments at depths $>$ 5 km. All segments of the Ortigalita fault appear to be skirting the edges of high-velocity bodies at depths $>$ 5 km in the vicinity of the pull-apart basin. Thus, while most strike slip faults located along the eastern margin of the Coast Ranges step left, the Ortigalita fault is the only one of these strike slip faults with a significant right step and associated pull-apart basin.
T31G-03 08:35h
Seismic Potential of a Compressional Inversion Province, NW South Island, New Zealand
In the NW South Island of New Zealand, large historical earthquake ruptures (1929 Murchison M7.7 and 1968 Inangahua M7.2) have dominant components of reverse slip. At the surface, active reverse faulting occurs along a segmented system of closely-spaced (10-20 km apart), N-S to NNE-SSW trending faults continuous over $>$ 150 km, and dipping 50-$80\deg$ both east and west. The faults truncate subparallel folds that deform the Tertiary sequence overlying a composite Paleozoic and Mesozoic crystalline basement. Some of the high-angle faults are inherited from Cretaceous extension, and have been subsequently inverted, consequent on right-lateral shearing and transpression on the Alpine Fault since the early Miocene. However, the deep geometry of these faults, their penetration into the basement and their relationship to the Alpine Fault are poorly understood. These problems have been addressed by reconstructing the structural contours at the base of the calcareous Oligocene sequence for the whole northwestern margin of the South Island. This represents a convenient strain marker, because: (i) it was deposited close to sea level; (ii) it is extensively preserved on land; (iii) it stands out as a high energy reflector in seismic lines; and, (iv) it has been penetrated by a number of oil exploration boreholes. Regional deformation of this Oligocene marker results from shortening episodes since the early Miocene with the development of sets of elongated and narrow N-S folds (wavelength $<$ 10 km). Crests of anticlines range in elevation from 1000 to 3000 m asl, whereas troughs of synclines are as low as -2000 m bsl. These large vertical displacements of the Oligocene marker are partly accommodated by reverse faulting (throws $<$ 3000 m), and partly by bedding-parallel slip along the steep flanks linking anticline-syncline pairs. Shortening calculated from section restoration is highly variable, ranging from $<$ 10%\ up to 30%\. These estimates represent a minimum because large-scale detachments of the cover sequence are likely to have been accommodated by blind thrusts that ramp across the crystalline basement and follow layer-parallel trajectories at the basement-cover interface, or within low competence horizons. The deformation of the whole region is indicative of large-scale bulk shortening of the sedimentary sequence, between two basement buttresses. The eastern basement block is bounded by the sub-vertical to steeply E-dipping Alpine Fault. The western margin is a complex basement high that can be followed all along the coast line, bounded by a segmented system of interconnected high-angle inverted normal faults and low-angle thrusts dipping both east and west. The overall style of deformation is suggestive of strong components of shortening to the west and in the footwall of the Alpine Fault, accommodated by inversion of high-angle normal faults in the basement, folding and detachment of the sedimentary covers, and propagation of new, low-angle thrust faults, that cut across the early structures and the detached folds. Nucleation of seismic ruptures in the basement at depths of $\sim$ 10-15 km may therefore lead to distributed deformation in the near-surface. Conversely, large seismogenic faults that have not ruptured in the past 200 years may be concealed beneath thick detached portions of the cover sequence.
T31G-04 08:50h
Microearthquake Distribution along the Itoigawa-Shizuoka Tectonic Line by the 2003 Temporary Array Observation
The Itoigawa-Sizuoka Tectonic Line (ISTL) is a major geological structure that divides Japan into NE and SW parts. It was formed as a normal fault in the early Miocene and represents the southwestern boundary of the northern Fossa Mangna sedimentary basin. During the Pliocene the ISTL is reactivated as a reverse fault due to tectonic inversion, after the collision of the Izu-Bonin arc with the Japanese arc. It comprises an active inland fault system with a high slip rate. The central and northern parts of the ISTL have a slip rate of 8.6-9.5 mm/yr and 3.0 mm/yr, respectively, while the southern part is considered to have ceased its activity. To reveal seismic activity, which may be related to the ISTL activities, we deployed seismic arrays in northern, middle, and southern parts. These arrays consisted of 4-9 stations in each area. At the southern part, across the ISTL, we also deployed a linear station array across the ISTL, consisted of 49 stations. We used 3 -component 1-Hz seismometers and long-term recorders with a sampling rate of 100 or 200 hz Hz. The observation lasted for 3 months starting from August 4 to November 21, 2003. In the north part of ISTL, using our temporary network and the local network in this area deployed by the Japanese Meteorological Agency (JMA) and the Earthquake Research Institute, the University of Tokyo, we determined 281 events with 2.7 $>$ M $>$ -1. The JMA network has located only 103 of these events. Most of which are concentrated in a cluster at a depth of about 15 km. In order to locate these events accurately, we have applied the Joint Hypocenter Determination method. We estimated station corrections for our temporary network and the regional permanent network. The main cluster is located inside an area of 1 x 1 x 1.5 km approximately. We can clearly see a migration in the hypocenter distribution from a starting depth of about 15.5 Km km to a depth of 14 km, which is much deeper than the deeper extension of the ISTL. Some previous studies in the area attributed this kind of cluster activity to water migration through cracks in the rock formations. The hydro-fracturing can be one of the triggering mechanisms of an inland large earthquake which occurs far away from a plate boundary.
T31G-05 09:05h
Structural geometry and seismicity of the Coalinga - Kettleman Hills blind-thrust system
The 1982 New Idria (Mw = 5.4), 1983 Coalinga (Mw = 6.5), and the 1985 Kettleman Hills (Mw = 6.1) earthquakes provide one of the clearest examples of damaging earthquakes occurring on a segmented blind-thrust system that is manifest at the surface in a series of en-echelon anticlines. These events lacked any surface rupture, but had uplift signatures associated with the anticlines, arguing for a direct relationship between blind thrust faulting, folding, and surface deformation. To investigate this relationship between coseismic folding and faulting, we present a new three-dimensional (3-D) structural model of the Coalinga - Kettleman Hills fold-and-thrust belt on the western border of the San Joaquin basin in central California. Using seismic reflection profiles, surface geology, well-bore data, and seismicity, we interpret the Coalinga anticline as being comprised of a stack of imbricated structural wedges, related to two major fault ramps at depth. These two forethrust ramps share a common upper detachment surface and merge with backthrusts to generate the gross morphology of the Coalinga anticline. The prominent forelimb of the Coalinga anticline that is expressed at the surface records slip on a fault that has branched off of this upper detachment surface. Based on our analysis, slip on the upper fault ramp causes folding of the Coalinga anticline, whereas displacement of the lower thrust ramp focuses deformation in a buried monoclinal limb that lies to the east of the mapped anticline. Based on hypocentral locations, the 1983 Coalinga (Mw = 6.5) mainshock ruptured the lowermost of these two ramps. To constrain the seismogenic portions of the structure, we use hypocenters from 1980 to 2002 that are relocated with L1-norm, waveform cross-correlation techniques. These relocated events illuminate a thrust ramp that strikes parallel to the axis of the Coalinga axis, but is significantly offset to the west from the hypocenter of the Kettleman Hills event. A segmented ramp of this nature is consistent with the model that a tear fault separates the ramps, which likely served as a rupture boundary for these events. In addition, moment density plots are employed to investigate how strain is accommodated in various structural compartments in the hanging wall and footwall of the thrust ramps.
T31G-06 09:20h
Seismicity in Thrust-and-Fold Belt of North Algeria: A Random or Stress Triggering Activity?
On May 21, 2003, a destructive earthquake (Mw=6.8) occurred near Algiers, the capital city of Algeria. The earthquake is associated with an offshore, NE-SW trending and 45° SE dipping thrust fault within the actively deforming Tell Atlas thrust-and-fold belt, part of the African-Eurasian plate boundary. In order to study the stress evolution and seismic hazard in the Algiers region, we apply Coulomb friction theory, using maps of surface ruptures and active faults, seismicity catalogue back to 19th century and well located aftershock distribution of the earthquake. Modeling of the historical and instrumental earthquakes in the past 100 years shows that static stress transfer does not seem to play an important role in triggering large earthquakes in this region. In other words, the earthquakes appear to be randomly located. However, it is not known whether this is a common case in a compressional tectonic setting or the time window used is too short to reveal stress interactions between large earthquakes. A longer catalogue may also answer how seismic activity in a thrust-and-fold belt evolves with time. By contrast with strike-slip faults (e.g. North Anatolian or San Andreas faults) that generally form a narrow deformation zone, thrust faults are diffused over a wide area within which fault segmentation is poorly understood. Taking into account the 1716 seismic sequence and fault-related fold segments near the epicentral area, the potential for a large earthquake in Mitidja Basin near Algiers has been increased by the May 2003 event.
T31G-07 09:35h
Contribution to a Detailed Seismic Hazard Assessment in the Lower Tagus Valley
Several previous works have assessed the seismic hazard in the Lower Tagus River area, an industrialized and populated area, SE of Lisbon, using historical and instrumental data. In this study we gave a contribution to a more detailed study of the seismic risk, and we have tried to clarify the epicentral features and the fault characteristics as well as the seismogenetic behavior. This is particularly difficult in this case because, although we have strong evidence of past strong events in the historical data, no such events is recorded in the instrumental data. We have reinterpreted previous seismic prospecting work in the area, as well as other geophysical data gather in the last years. We believe that the reinterpretation of the seismic profiles, near the villages of Vila Franca and Azambuja, allows an identification of the seismogenetic structure, close to the surface, while some of the natural seismic events recorded seem to be generated by the deepest part of the same structure. This interpretation is with good agreement with the gravimetry data. We have also be able to calculate the probability of the top acceleration to exceed a certain value in a given time intervals, in each point of the area.