T21B-1940
Very High Resolution Optical Images for Detecting Co-seismic Surface Effects: the Cases of the 2005 Kashmir (Pakistan) and the 2003 Bam (Iran) Earthquakes
Very High Resolution (VHR) satellite panchromatic image has revealed to be a reliable tool to detect surface effects of natural disasters. This is particularly true whereas the hit territory is a remote land and/or with logistic and security problems. Data from this kind of sensor have a potential for more exhaustive and accurate mapping of the environment with details of sub-meter ground resolution. We show two large earthquake case studies, the 2005 Mw 7.6 Kashmir and the 2003 Mw 6.6 Bam events, both producing significant surface effects as ruptures, landslides and building damages. In order to test the capability of VHR images to recognize and evaluate such features we used panchromatic QuickBird imagery (0.6 m spatial resolution) acquired before and after the events (kindly provided by DigitalGlobe). Concerning the Pakistan we focus on the Muzaffarabad and Balakot areas, both crossed by the earthquake fault and experiencing edifice collapses. Same sort of analysis is performed for the ancient town of Bam. We proceed with: 1. identification on the images of the main rupture trace and of major landslides; 2. generation of a detailed spatial distribution of damage and collapses through a single building automatic classification approach; 3. cross-comparison of the different surface effects. The QuickBird panchromatic images provide a view of the co-seismic features at large scale, revealing complex geometric pattern of the cracks and compressional deformation features. It is possible to detect the lateral sense of movement, and based on the sun shade projection in the images, we infer the facing of the scarp, thus the uplifted side. Regarding point two, if in one hand the use of QuickBird images leads to detect very small details, on the other hand buildings become rather complex structures. Furthermore they may be surrounded by scattering objects making less evident the contrast between the roofs and the ground, thus increasing the difficulties in the classification process. This implies that a single band is not enough to reduce false alarm signals and classify such complex environment. Therefore other information such as the shape of objects is extracted using a multiscale textural analysis, based on mathematical morphology filters.
T21B-1941 [WITHDRAWN]
Paleoseimic investigations along the Ganos fault and characteristics of the 9 August 1912 Ganos earthquake (Ms=7.3) - North Anatolian Fault / Turkey
The Ganos fault is the westernmost segment of the North Anatolian Fault (NAF), limited by the Marmara Sea and the Aegean Sea on both ends. Active deformation of the right-lateral strike-slip fault is clearly expressed by several small pull-aparts and sag ponds, pressure and shutter ridges and offset streams along the 45 km long onland section. The most recent destructive earthquake along this section is the Ms = 7.3 Mürefte earthquake of 9 August 1912. The 45-km-long co-seismic 1912 surface ruptures and related displacements have been investigated to document further the slip distribution. Right lateral offsets of roads, paths, streams, man-made buildings and field limits have been measured at 40 sites using differential GPS and total station. The offset distribution ranged from 2 to 5.5 m with a maximum of 5.5 m near Gaziköy. We conducted paleoseismic investigation on three sites along this part of the NAF. Seven trenches were excavated at the Güzelköy site, which provided evidence for three earthquakes since the 11th century with cumulative right-lateral offset of 12.5 m and 23.6 m on a paleo-channel. Radiocarbon dating of the paleo-channel implies a slip rate of 17.5 - 20 mm/year on this section of the fault. Comparison of dating results with historical earthquake catalogs point towards the 1343, 1659 or 1766 and 1912 earthquakes in the Ganos area. Preliminary results of the Yeniköy and Yörgüċ trench sites present evidence for three and two or three past earthquakes, respectively. Faulting related with the 1912 Mürefte earthquake are visible in both the Güzelköy and Yeniköy trenches. Other trenching studies provide further evidence of faulting next to the Saros bay. Field observations suggest that the entire on-land section of the Ganos Fault ruptured during the 1912 earthquake. Recent submarine earthquake studies propose an offshore fault continuation of 40 to 60 km to the east of Gaziköy, into the Marmara Sea basins. Paleoseismic studies, slip distribution, fault segmentation and the earthquake size suggest 100 - 120 km-long surface rupture for the 1912 earthquake, with 2.5 - 3 m average slip. Our results infer the maximum 1912 rupture length and propagation and limit the slip deficit of the seismic gap in the Sea of Marmara.
T21B-1942
Defining Additional Stratigraphy in Paleosismic Trenches by 2D Logging of Magnetic Susceptibility. A Paleoseismic Investigation Near Lake Ladik, North Anatolian Fault, Turkey.
The North Anatolian Fault (NAF) is a dextral strike-slip plate-boundary fault zone extending ~1400 km in an arc across northern Turkey. We opened a paleoseismic trench ~2.7 km NW of Destek village on a segment which ruptured (for ~280 km) in the 1943 Ladik Earthquake (Mw:7.7). Sediments exposed in the trench yielded information on the timing of at least 6 paleoearthquake events during the last 3000 years in addition to evidence of the 1943 event. The trench was excavated across an uphill-facing fault scarp caused by an overturned thrust fault splay of the NAF near a localized restraining bend. The uphill-facing scarp trapped sediment derived from a small (~2 ha), non-channelized catchment with erosion in the steeper upper half and deposition in the lower half. Conventional descriptive trench logging of the, southern, up-thrown side of the fault identified weathered rock of various lithologies, grading upwards to residual soil exclusively on the west trench wall, and capped by topsoil on both walls. On the down-thrown, northern side of the fault, we logged a sequence of colluviums and poorly defined paleosols. Conventional trench logs alone do not negate the possibility that the stratigraphy observed north of the fault formed due to climate cycles or anthropogenic processes, rather than earthquakes. Magnetic susceptibility (MS) measurements provide a link between the sediment trap strata and fault rupture. Two-dimensional logging of magnetic susceptibility, using a MS2E Bartington point sensor, was undertaken on the west wall of the trench. The residual soil on the up-thrown side of the fault displayed low MS values overlying rock with relatively high MS values which we interpret as the result of leaching by supergene processes. Wedges of low MS values were identified on the down-thrown side of the fault in a soil of otherwise intermediate MS values representing colluviums sourced from the catchment. The low MS wedges are interpreted to be colluviums derived from the residual soil on the up-thrown block due to collapse of the fault scarp following fault rupture. The presence of the low MS wedges helps to define a sequence genesis model whereby paleosols are buried following earthquakes. Without the magnetic susceptibility data the link between the cyclic sedimentation pattern and earthquake cycles would be tenuous, relying solely on correlation with known earthquakes.
T21B-1943
Tectonics and Tsunami Hazard in the Eastern Mediterranean Inferred From the AD 365 Earthquake
Historical accounts describe an earthquake and tsunami in the Eastern Mediterranean on 21 July AD 365 that destroyed cities and drowned thousands of people in coastal regions from the Nile Delta to modern-day Dubrovnik. It has long been suggested that the earthquake was caused by slip in the Hellenic subduction zone near Crete, where relict shore lines raised up to 10~m above sea level have been recognised since the 19th century. We present evidence from radiocarbon dating and field observations that these shorelines were uplifted synchronously with the AD 365 earthquake. The distribution of uplift, combined with present-day seismicity, suggest that this earthquake did not occur on the shallowly dipping subduction zone interace beneath Crete, but on a more steeply dipping fault in the overriding plate. Such an earthquake would cause uplift of the sea floor, and calculations of tsunami propagation show that the resulting tsunami would have travelled through most of the Eastern Mediterranean, reaching open ocean wave heights comparable to those following the 2004 Sumatran earthquake. GPS measurements of crustal shortening in the region of Crete yields an estimate of ~5000 years for the repeat time of such earthquakes on this single fault, an analysis supported by radiocarbon dates on uplifted marine terraces. If the same process takes place along the whole 600~km long subduction zone, these events may repeat once every 800 years.
T21B-1944
Peculiar Active-Tectonic Landscape Within the Sanctuary of Zeus at Mt. Lykaion (Peloponnese, Greece)
The Sanctuary of Zeus (Mt. Lykaion) lies in the Peloponnese within the Pindos fold and thrust belt. It is the object of investigation of the Mt. Lykaion Excavation and Survey (http://lykaionexcavation.org/). Mt. Lykaion is a thrust klippe, on the summit of which is an upper sanctuary marked by an ash altar, temenos, and column bases. Earliest objects recovered from the ash altar go back to 3000 BCE, leading Dr. David Romano (University of Pennsylvania), a principal leader of the project, to conclude that worship of divinities on the summit is ancient. Detailed structural geological mapping reveals one dimension of the "power" of the site. Crisscrossing the upper sanctuary are scree bands that mark the traces of active normal faults, which are expressions of tectonic stretching of the Aegean region. The scree bands, composed of cinder-block-sized limestone blocks, range up to 10 m in outcrop breadth, 100 m in length, and 5 m in thickness. Though discontinuous, most of the scree bands lie precisely on the traces of through-going faults, which cut and displace the sedimentary formations of the Pindos group. Some cut the thrust fault, whose elliptical trace defines the Lykaion klippe. What makes the scree bands of this active-tectonic landscape "peculiar" is that there are no cliffs from which the scree descends. Rather, the bands of scree occur along flanks of smooth, rounded hillslopes and ridges. The scree bands coincide with modest steps in the topography, ranging from tens of centimeters to several tens of meters. The specific bedrock formation where the bands are best developed is an Upper Cretaceous limestone whose average platy-bedding thickness (approximately 20 cm) matches closely the average joint spacing. The limestone has little mechanical integrity. It cannot support itself as a scarp footwall and instead collapses into a pile of scree, whose upper-surface inclination conforms to a stable angle of repose. Evidence of the contemporary nature of this faulting includes a scree band that nearly completely covers stone structures built by shepherds. Though the scree bands conceal surface ruptures, it is expected that trenching will determine that the scree cover may have preserved beneath it some expressions of surface rupture, and perhaps fault surfaces themselves.
T21B-1945
Archaeoseismology of the Halai Neolithic Site, Central Greece
Greece has been continuously inhabited since ancient times, providing a rich archeological record with which to examine the potential impacts of geological processes. Greece is a seismically active region undergoing crustal extension due to subduction roll back at the Hellenic Trench. By identifying seismic deformation of archaeological remains and accurately mapping the tectonic features of the surrounding region, it has been possible to better estimate the activity of large faults that have not ruptured within the brief window of historic seismic observation. The Halai archaeological site in central Greece contains archeological remains that record eight thousand years of human habitation. The site is located within 6 km of large, potentially active normal faults. Evidence for major seismic deformation is found in the oldest buildings at Halai, and likely indicates one or more major earthquakes between the Late Neolithic and Archaic periods on the nearby Atalanti or Arkitsa faults. Further evidence for Holocene deformation comes from geological mapping of the faults. The Arkitsa and Atalanti faults offset bedrock against unconsolidated sedimentary deposits, have fault scarps marked by topographic lineaments, and define triangular facets at the range fronts. Normal faults offset unconsolidated sedimentary deposits at two locations observed during fieldwork. One hundred and sixty-eight measurements of normal faults were used to estimate the principal stress axis. Comparison of the fault data with earthquake focal mechanisms shows that both datasets record north-south extension. The combined dataset indicates the Atalanti and Arkitsa faults represent a long-lived seismic hazard in central Greece that has affected people since at least Neolithic times.
T21B-1946
Geological and Macroseismic Data For Seismotectonic Purpose: The 1706 Maiella (Abruzzo, Italy) Earthquake Case Study
The nature and distribution of the seismicity and of the active structures in central Italy show that the active deformation field is mainly characterised by extension in the axial zone of the Apennines and by co-axial contraction on the frontal part of the belt. In this tectonic context become crucial, from the seismic hazard point of view, the seismotectonic characterization of the major earthquake localised between the two seismotectonic provinces. The 1706 (Io=IX-X), 1933 (Io=VIII-IX) and 1881 (Io=VIII) Maiella earthquakes stroke areas extending outward of the easternmost, NNW-SSE, active normal fault alignment and inward of the N-S oriented active thrust front. These earthquakes have been attributed by some authors to thrust faulting, while partly to upper crust normal faulting and partly to thrust faulting by others. Due to the poor local configuration of the national seismic network, the available seismic instrumental data are inadequate to constrain the active deformation pattern of the Maiella area. On the other hand, the shallow and deep tectonic setting is rather well known and macroseismic data of the afore-mentioned earthquakes available. The present study has been carried on mainly focusing on the 1706 event, following and integrating three methodological steps: a) selection and definition of the likely 3D seismogenic source models; b) evaluation of possible local effects on the macroseismic field data; d) estimation of seismic scenario in terms of macroseismic intensity, calculating synthetic strong motion time histories starting from different configuration and depths of the seismogenic source models, with a stochastic finite-fault modelling of the ground motion. The method involves discretization of fault plane into smaller sub- faults; the contribution from all the sub-faults is summed to produce the synthetic acceleration time history. For each point of the 1706 macroseismic field, peak ground acceleration and velocity were determined in order to evaluate the calculated intensities using empirical relationships. Results show that the seismic scenario closest to the 1706 observed intensity field may be reproduced starting from a source model associated to SW-dipping thrust faulting located at depths between 5 and 10 km.
T21B-1947
Morphotectonics of LANF-controlled Plio-Quaternary grabens (Lunigiana-Garfagnana, Italy)
We report first results of structural and morphotectonic analyses aimed to characterize active and potentially seismogenic faults in the Lunigiana and Garfagnana basins. These basins are Pliocene-Holocene asymmetric grabens that lie in the hanging wall of a regional, east-dipping, low angle detachment that is imaged in seismic profiles deepening beneath the Apennines down to a depth of ~13 km. Seismic lines highlight E-dipping synthetic faults dipping between 30° and 60° down to depths of ~5km before soling onto the low-angle detachment. In contrast, W-dipping antithetic faults show on average higher dip angles of 50° to 70° and root on the detachment at depths of less than 5km. Topographic metrics including drainage patterns, river long profiles, indices of channel concavity, stream-length gradient of modelled long profiles, steepness and swath profiles are used to better constrain the effects of active slip on these faults as they rupture and help shape the current topography. We quantify the relationships between these faults and watershed-scale geomorphology using a 10 m digital topography to extract channel and basin metrics. The E-dipping faults with strongest geomorphic signature of Late Quaternary activity are the Mulazzo and Olivola-Soliera faults in the Lunigiana and the Casciana-Sillicano and Bolognana-Gioviana faults in the Garfagnana. The most important W-dipping splays in terms of late Quaternary activity are the Groppodalosio and Compione-Comano faults in the Lunigiana and the M.Prato-Colle Uccelliera-M.Mosca alignment in the Garfagnana. The Tendola-Equi Terme-Gramolazzo alignment is a high angle E-W right- lateral transfer zone. This alignment, quite evident in aerial photographs and in seismic lines, is interpreted as a transfer zone of active extension between the Lunigiana and Garfagnana grabens. Computation of geomorphic indices is in good agreement with the seismological data, suggesting that historic earthquakes in the area can be associated with these mapped faults that are further capable of generating surface ruptures and contributing to the production of local relief.
T21B-1948
Late Pliocene To Pleistocene Tectonic Activity In SW Portugal: The S.Teotónio-Aljezur- Sinceira Fault System And Evidence For Coastal Uplift
Southwestern Portugal is located close to the Eurasia-Nubia plate boundary. East of the Gloria transform fault, this boundary becomes complex, particularly as it approaches the Gorringe Bank, the Horseshoe Plain, and the Gulf of Cadiz, where deformation related to the NW-SE convergence of Iberia and Nubia, at a rate of ~4-5 mm/ year, becomes distributed across a few hundred kilometer-wide zone. This area corresponds to the inferred seismogenic source zone for the 1755 earthquake and tsunami (estimated ≥ Mw 8), and also for the Mw 7.9 1969 event. During the past decade, several off-shore active folds and faults have been recognized in this region however, in spite of increased knowledge, none of the recognized active structures are clearly associated with the 1755 earthquake. Major likely sources are the Marquês de Pombal and Horseshoe faults. The Marquês de Pombal fault is a major NNE-SSW trending thrust located ~100 km SW of Cape S.Vicente that exhibits a ~1 km-high, 60 km-long scarp. Assuming rupture of this entire structure suggests earthquake magnitudes in the Mw 7.8 range. The Horseshoe fault, which is oriented NE- SW along a 175 km-long trend parallel to Säo Vicente canyon, a major morphological feature in the off- shore that has been interpreted as a possible extend for the Alentejo-Plasencia fault. Rupture of this entire fault could yield moment magnitude events up to Mw 8, assuming 10 m of average displacement. Neither of these potential sources can likely produce, by themselves, an earthquake that matches the upper estimates for the 1755 earthquake (Mw 8.7). Along the southwestern Portuguese coast, mainly at the western coastline, cliffs in Palaeozoic schist reach more than 100m in altitude, with evidence of uplift in the form of raised beach deposits, paleo-sea cliffs and multiple eolianite units. Several abrasion platforms with regional expression may have formed during multiple marine occupations. In contrast, the southern coast is underlain mainly by Mesozoic limestones and exhibits a generally lower topography accented by karstic morphology. In both areas, little work has been done to map the sequence of marine terraces, nor to determine their ages, although the majority of them are likely Pleistocene. The highest raised marine deposits reach an altitude of 370 m ~13 km inland and may be as old as Pliocene in age. Inland, the Säo Teotónio–Aljezur–Sinceira fault system (STASFS) extends NNE-SSW for 50 km, parallel and close to the southwest Portuguese coast, and controls the development of several small Cenozoic tectonic basins. It comprises onshore faults which may relate to the ongoing plate boundary deformation. This fault system expresses primarily sinistral strike-slip with a minor reverse component. Four cenozoic strike-slip basins occur along the STASFS, generally with lengths of less than 5 km and a maximum width of 1.5 km, filled with Miocene to Pleistocene sediments. In some areas, fault-related post-Pliocene vertical displacements of up to 100 m may have occurred, but generally they only reach a few tens of meters. This coastal region is therefore particularly appropriate for establishing the offshore-onshore link through a detailed neotectonic study of the active faults, including exploration with paleoseismological techniques, and the vertical deformation field using marine terraces as a reference frame.
T21B-1949
Assessing the recurrence of big earthquakes and tsunami in the Gulf of Cadiz (SW Iberia) using thin-sheet neotectonic modeling
The eastern end of the Azores-Gibraltar plate boundary is characterized by distributed deformation that accommodates the collision between the Eurasia and Africa plates. Despite this, the active faults in the area can generate very large earthquakes and destructive tsunamis, such as the Great Lisbon Earthquake, that occurred in the 1st November 1755 (estimated Mw = 8.7). The largest instrumental earthquake recorded was the 28th of February 1969 Mw=8.0, localized in the Horseshoe abyssal plain with a thrust fault mechanism. In this study we used a thin-shell approximation (SHELLS- Bird, P., Computers and Geosciences, 25, 383– 394, 1999) to model the neotectonics of this segment of the Africa-Eurasia plate boundary and put constraints on the recurrence periods of earthquakes and tsunamis. In relation to previous neotectonic models in the region we use a better constrained structural map, particularly in the Gulf of Cadiz and SW Iberia, based on recently acquired multi-beam bathymetry, backscatter data and numerous high quality multi- channel seismic profiles. Importantly, the map shows the existence of several NNE-SSW to ENE-WSW thrust faults, associated to prominent bathymetric features, and a set of very long (up to 600 km) strike-slip lineaments, extending between the western Horseshoe Abyssal Plain and the eastern Gulf of Cadiz. Different models have been tested, for various boundary scenarios (geometry and plate velocities) and fault networks, and the results compared with seismic strain release, recent GPS observations and stress orientation. The modeling suggests that, when mature, the long strike-slip lineaments will accommodate most of the relative motion between Eurasia and Africa (aprox. 4 mm/a) along a "transform-type" plate boundary. This situation, however, is associated with only minor thrust faulting in the region and predicts a strong attenuation of the velocity field between the northern Morocco and Gibraltar, in contradiction with present day GPS measurements. In our preferred tectonic model, the strike-slip lineaments appear as segmented features and a significant amount of the Africa-Eurasia plate convergence (1-2 mm/a) is accommodate along the NE-SW thrust fault systems located in the northern Gulf of Cadiz and SW Iberia, probably linked through NE-SW transfer faults. Accordingly, several large active faults can generate earthquakes with a magnitude greater than 8.0 and an overall recurrence period lower than 1000 years. For the very large, "1755-like" earthquake and tsunami, the thin-sheet modeling results imply a recurrence interval of 10 000 years.
T21B-1950
Is Downtown Seattle on the Hanging Wall of the Seattle Fault?
The Seattle fault is an ~80-km-long thrust or reverse fault that trends east-west beneath the Puget Lowland of western Washington State, and is interpreted to extend beneath the Seattle urban area just south of the downtown area. The fault ruptured about A.D. 930 in a large earthquake that uplifted parts of the Puget Sound shoreline as much as 7 m, caused a tsunami in Puget Sound and extensive landslides throughout the area. Seismic reflection profiles indicate that the fault has 3 or more fault splays that together form the Seattle fault zone. Models for the Seattle fault zone vary considerably, but most models place the northern edge of the Seattle fault zone south of the downtown area. These interpretations require that the fault zone shifts about 2 km to the south in the Seattle area relative to its location to the east (Bellevue) and west (Bainbridge Island). Potential field anomalies, particularly prominent magnetic highs associated with dipping, shallow conglomerate layers, are not continuous in the downtown Seattle area as observed to the east and west. Compilation and re-interpretation of all the existing seismic profiles in the area indicate that the northern strand of the Seattle fault, specifically a fold associated with the northernmost, blind fault strand, lies beneath the northern part of downtown Seattle, about 1.5 to 2 km farther north than has previously been interpreted. This study focuses on one previously unpublished seismic profile in central Puget Sound that shows a remarkable image of the Seattle fault, with shallow subhorizontal layers disrupted or folded by at least two thrust faults and several shallow backthrusts. These apparently Holocene layers are arched gently upwards, with the peak of the anticline in line with Alki and Restoration Points on the east and west sides of Puget Sound, respectively. The profile shows that the shallow part of the northern fault strand dips to the south at about 35 degrees, consistent with the 35 to 40 degree dip previously interpreted from tomography data. A second fault strand about 2 km south of the northern strand causes gentle folding of the Holocene strata. Two prominent backthrusts occur on the south side of the anticline, with the southern backthrust on strike with a prominent scarp on the eastern shoreline. A large erosional paleochannel beneath west Seattle and the Duwamish waterway extends beneath Elliot Bay and obscures potential field anomalies and seismic reflection evidence for the fault strands. However, hints of fault-related features on the profiles in Elliot Bay, and clear images in Lake Washington, indicate that the fault strands extend beneath the city of Seattle in the downtown area. If indeed the northern strand of the Seattle fault lies beneath the northern part of downtown Seattle, the downtown area may experience ground deformation during a major Seattle fault earthquake and that focusing of energy in the fault zone may occur farther north than previously estimated.
T21B-1951
The western extension of the Seattle fault: new insights from seismic reflection data
Seismic reflection profiling across the Seattle fault west of Puget Sound, western Washington State, reveals near-surface deformation as far west as Hood Canal. North of Green Mountain and in line with the trend of the Seattle fault, folding within late Pleistocene sediments and a deflected stream channel indicate active deformation. Here, we observe folding in late Pleistocene deposits overlying more steeply dipping Tertiary strata previously inferred to be part of the Seattle monocline. The greater than 70 km long dominantly east- west striking Seattle fault follows the contours of the northwest edge of Green Mountain and locally strikes south-southwest, or a ~25 degree rotation in local fault trend. Exposures of Eocene bedrock (Crescent Fm) south of the fault indicate deeper exposures than along the eastern portions of the Seattle fault. This deeper exposure, accompanied with greater degrees of deformation than observed to the east, suggest higher uplift rates above the Seattle fault west of Puget Sound. Seismic profiles west of Green Mountain, along strike with the Seattle fault west near the west end of the Seattle Basin, show a complex pattern of late Quaternary and older deformation. However, these profiles extend beyond the inferred limits of the Seattle Basin and do not show the characteristic monocline and deformation front that defines the Seattle fault to the east. A west-northwest striking fold within Late Quaternary sediments, steeply dipping Tertiary strata, and a southwest-striking magnetic lineament all suggest a complex and possibly distributed pattern of deformation along the west part of the fault. The Seattle fault continues west beyond its mapped extent north of Green Mountain; it may merge with structures that control the Dewatto Basin along the south edge of the Seattle uplift, including the Frigid Creek or Saddle Mountain faults west of Hood Canal.
T21B-1952
Segmentation of the Tacoma fault: evidence from seismic and magnetic data
We integrate seismic and magnetic data to characterize the Tacoma fault and northern portions of the Tacoma Basin in western Washington State. The Tacoma basin, a deep (greater than 4 km) basin filled with Eocene and younger sediments, is a major physiographic feature underlying southern Puget Sound. Its northern margin is bounded by the Tacoma fault, an east-striking, north-dipping backthrust that forms the southern margin of the Seattle uplift and accommodates north-south shortening of the Puget Lowland. Whereas the east-striking Seattle fault extends >70 km with a consistent style of deformation, we find evidence that the Tacoma fault is segmented by north-northeast-trending faults. A prominent east-striking magnetic lineament coincides with lidar-identified fault scarps and defines the western part of the active Tacoma fault. This lineament extends 15 km from Hood Canal east to Key Peninsula and is characterized in seismic profiles as a narrow, north-side-up kink-band. Land and marine seismic profiles, and a change in trend of magnetic lineations near Key Peninsula suggest that the Tacoma fault either terminates at this location or significantly changes character. Shortening east of Key Peninsula is expressed, in part, by the Rosedale monocline, a southeast-trending structure that shows evidence for active uplift. The Rosedale monocline extends ~20 km to the eastern margin of Puget Sound, where seismic profiles show a second north-northeast-trending set of faults. Land seismic profiles and magnetic data east of Puget Sound are consistent with the Tacoma Basin shallowing near the eastern boundary of Puget Sound. The apparent synchronous timing of large earthquakes approximately 900 AD on both the Tacoma and Seattle faults suggest these fault systems are kinematically linked, yet the surface and subsurface expression of the two fault systems are distinct. Based on the fault segment lengths, our findings suggest the Tacoma fault can support earthquakes < M 6.5 and that no backthrust on the Seattle fault system is evident in the upper one km on our data east of Puget Sound. Thus, if fault ruptures are restricted to individual segments, hazards along the Tacoma fault may be less than previously modeled.
T21B-1953
Comparison of geodetic and paleoseismic rates of deformation in the Puget Sound- Georgia Basin, Pacific Northwest
We compare geodetic rates of horizontal shortening in the Puget Sound-Georgia Basin region with long-term rates of horizontal shortening calculated from paleoseismic studies across the same region. Our region of comparison encompasses a zone of concentrated seismicity, lying within a rectangle centered at W122.5° and N48°, with edges defined by ±1.5°. Within this region, geodetic rates are well constrained by a network of permanent Global Positioning System (GPS) sites. The long-term N-S shortening rate is 3.2±0.8 mm/yr (after subtracting interseismic subduction loading). Extending this rate back though the Holocene yields between 24 and 40 meters of horizontal shortening to account for with paleoseismic investigations. Paleoseismic studies on 11 fault zones within our region of interest provided slip estimates of past earthquakes. These fault zones include five thrust, two normal, and four oblique faults. Paleoseismic shortening rates are calculated using stratigraphic offsets and measurements of fault dip. Stratigraphic offsets consist of two types of measurements: 1) vertical separations of strata by faults observed in excavations, and 2) evidence for coastal uplift and subsidence. Fault dips are estimated from published reports (e.g., seismic reflection lines) or directly measured in excavations. Minimum offsets are measured in almost all cases. Using these techniques, we can account for a minimum of about 36 meters of horizontal shortening during the Holocene. Uncertainties in the estimation of paleoseismic shortening center around measurement of fault dip, slip per event, and missing events in the record. Fault dips at the surface are measured accurately from scarp excavations and seismic sections and likely result in only a small amount of error. Slip per event is more difficult to estimate because in a few cases, there may be significant out of plane motion that cannot be accurately measured. We think that the record of large earthquakes is fairly complete within the study area based on thorough investigations using a variety of geophysical techniques to identify possible fault zones, including Lidar, aeromagnetic surveys, seismic reflection, and tomography. Paleoseismic studies on faults identified by high-amplitude geophysical anomalies and on scarps identified through analysis of lidar data have likely found the majority of the largest earthquakes. We know of two ongoing studies that identified evidence for Holocene activity that we have not accounted for, and may add an additional three to four meters to our estimate. Within the Puget Sound-Georgia Basin region, long-term rates of horizontal shortening estimated using GPS compare favorably with long-term estimates of Holocene shortening. Shortening estimated from the paleoseismic record accounts for about 100% of the total Holocene rate suggested by GPS, within the uncertainties of each technique. About 50% of the total Holocene shortening occurred during large earthquakes on several faults around 1100 years ago, at or about the same time as the penultimate Cascadia subduction zone earthquake. This supports the idea that abrupt N-S shortening in the Puget Sound-Georgia Basin region follows great subduction zone earthquakes and could trigger earthquakes on shallow crustal faults like the Seattle fault if the faults are already near failure equilibrium.
T21B-1954
Crustal Strain Rate Tensors From Dense GPS Networks
Some of the most destructive seismic events such as the 1994 Northridge MW 6.7 and the 1995 Kobe MW 6.8 earthquakes showed in a dramatic way the shortcoming of traditional seismic hazard assessment, which is mainly based on the recorded seismic history. As recent satellite based geodetic measurements like GPS and InSAR provide insights of crustal motion and deformation at new scales, both in accuracy and spatial and temporal resolution, they are of increasing importance for seismic hazard assessment. In particular, with the deployment of dense networks like the one of SCEC in southern California and others in different parts of Europe, southern America and Asia. However, in order to characterize active faults in terms of slip rates, locking depths, and earthquake recurrence times the geodetic data have to be interpreted using models and can be biased by the implicit assumptions. Furthermore a preliminary knowledge of the location of the active faults is often required for the modeling, increasing the difficulty of identifying unknown active structures. To avoid these uncertainties we apply geostatistical interpolation schemes to interpolate geodetic data and to obtain a continuous velocity field for each component. These fields are the basis for continuous (time dependent) strain rate tensors, which provide information like the spatial distribution of strain rates, direction of maximum shear strain rate, dilatation, etc. In regions with dense geodetic networks (e.g. Southern California and Western China) we can easily identify areas of high strain often associated with active faults, and estimate the errors introduced by the interpolation. Our interpolation scheme, that can easily be applied wherever geodetic data are available, clearly identifies and characterizes the active fault systems. For example in the eastern margin of the Tibetan Plateau our results are compatible with strain rate fields by Liu and Bird (2008) obtained by finite element interpolation of earthquake focal mechanism. Our results can be utilized as a starting point for further numerical models and/or geological investigation to estimate current activities of faults.
T21B-1955
A Statistical Approach for Comprehensively Understanding Crustal Activity in Terms of Seismicity and Strain Rate
For comprehensively understanding crustal activity, we statistically investigated the temporal changes in relationships between geophysical measures for strain rate (dilatation rate and maximum shear strain rate) and seismicity (seismic energy and the number of earthquakes). For this research, we transformed a dataset of daily coordinates of GPS stations operated by Geographical Survey Institute, Japan, and the hypocenter catalog unified by Japan Meteorological Agency into the datasets for strain rate and seismicity with a same high-resolution grid format, respectively. Spatiotemporal relationships between the geophysical measures were statistically examined by drawing scatter diagrams using the transformed datasets for seismicity and strain rate with a same grid format in regions surrounding the source areas of some representative large inland earthquakes such as the 2003 northern Miyagi prefecture earthquake, and the 2004 Mid-Niigata Prefecture Earthquake. In result, release of larger earthquake energy was initiated by relatively smaller- magnitude earthquakes prior to the large earthquake in areas of smaller strain rates rather than of larger ones. This result was also compared with the temporal changes in other newly developed measures: the mean of each of the earlier four measures for a circular region centered at the epicenter of a large inland earthquake, that for a broader circular region, and the ratio of the former to the latter.
T21B-1956
Coseismic fault model of the 2008 Iwate-Miyagi Nairiku earthquake deduced from a dense GPS network
@The 2008 Iwate-Miyagi Nairiku (inland) earthquake (hereafter IMEQ) occurred beneath the border between the Iwate and Miyagi prefectures in northeastern (NE) Japan. Based on continuous GPS measumrements, Miura et al. (2002, 2004) found the existence of a notable strain concentration zone with an east-west (EW) contraction along the Ou Backbone Range (OBR), where the Volcanic Front runs through. This EW contraction zone also shows a higher activity of earthquakes occurring mostly in the upper crust, and includes the focal areas of large inland earthquakes. The main shock and its many aftershocks of the IMEQ were located within this tectonically active region. Furthermore, an active fault named the Kitakami- Teichi Seien Fault Zone (KTSFZ) exists close to the IMEQ focal area. The KTSFZ is composed of several small active faults and the southernmost one is called the Dedana Fault (DF). We aggregate the continuous GPS stations from four different institutions, including Tohoku University (TU), Geographical Survey Institute (GSI), Japan Nuclear Energy Safety Organization (JNES), and National Astronomy Observatory of Japan (NAO), for estimation of the coseismic fault model estimation. The estimated model explains the data well and reproduces the complex spatial pattern of surface displacement, especially in the northern part of the focal area. The estimated amount of moment release is equivalent to Mw 6.9, and the maximum slip reaches 3.5 m on the southern sub-fault. The JNES sites provide a strong constraint to estimate the coseismic fault parameters. The location of the upper edge of the model fault does not coincide with the surface trace of the DF. The GPS data suggests that the upper edge of the coseismic fault is located several kilometers west of the surface trace of the DF. Thus we can conclude that the IMEQ occurred on an unidentified fault system other than the DF.
T21B-1957
Relationship between the recurrence interval of fault activity and the thermal gradient in Japanese inland area
Information on the recurrence interval of an inland active fault is valuable for not only disaster prevention planning but also the scientific aspect. Recently, promoted by the Headquarters for Earthquake Research Promotion and other organizations in Japan, vigorous studies on inland active faults have been made. The results enable us to refer many kinds of information on the fault activity including the recurrence intervals of individual active faults. However, since it is difficult to infer an accurate history of the activity of a fault, the uncertainty of the recurrence interval is very large in many cases. If we can use a certain physical parameter which represents the characteristics of the crust to estimate the average recurrence interval of a fault, the situation will be much improved. We here discuss the usefulness of geothermal gradient data for the estimation of the accurate recurrence interval. In order to evaluate the utility of the geothermal data, we examine a relationship between the recurrence interval and the geothermal gradient in Japanese inland area. The spatial distribution of the geothermal gradient is estimated by low (longer than 100km)-pass-filtering of the gradient data observed in deep (deeper than 1000m) boreholes. As the results, it is suggested that the recurrence interval has a tendency to be short in the region where the geothermal gradient is high. Moreover, there is a tendency that the recurrence interval of a large fault is short. Combination of these relations can be used for the estimation of the earthquake recurrence interval.
T21B-1958
Ocean bottom gravimetry near the coast
Inland earthquakes have recently caused damages in the Japanese Islands. Although more than 2,000 active faults have been recognized in and around Japan, the earthquakes had their sources in unexpected faults in the upper crust. Detecting such hidden active faults has become an important subject in the field of earth sciences. Precise gravimetry, a basic geodetic measurement, is now recognized as one of important means for detecting active faults hidden under the ground. Precise and dense gravimetry has covered the Japanese Islands, but not in the offshore area. The most critical point is that there are no gravity data in the offshore area near the coast; the area has not been included in land nor marine gravity mapping. An ocean bottom gravimeter (OBG) was developed in Earthquake Research Institute, University of Tokyo, for gravimetry on the deep seafloor (Fujimoto et al., 1998). Free gimbal suspensions with an oil damper keep a sensor package of Scintrex CG-3M/SB gravimeter roughly vertical, typically in a few tens arc-seconds, and the effect of the remaining tilt is numerically corrected by the gravimeter. The gimbals and the sensor are in a pressure-tight spherical housing made of titanium alloy, and the logging unit in a glass sphere. Two trial measurements on the shallow seafloor in Suruga Bay showed that precision of seafloor gravimtery with the OBG is much the same that with field gravimeters on land (Fujimoto et al., 1998). The importance of gravity mapping over the shallow ocean bottom near the coast was recognized and there were requests to use our OBG for such observations. Then we modify the OBG system into a single housing for easy maintenance and measurement. Ocean bottom gravity measurements with the OBG have been carried out around Awaji Island near Kobe, off the northern end of Tohoku, off Niigata, and in the Seto Inland Sea. The OBG was lowered from a boat to each gravity point on the bottom with a rope. Judging from the observed data, precision of gravimetry depends on the condition of the OBG on the bottom. Unless the rope disturbed the OBG, the performance was much the same with gravimetry on land. The observed data and operation for the measurements will be reported.
T21B-1959
Age and Slip Distribution of Past Earthquakes Along the Bogd Fault (Mongolia)
We carried out morphotectonic and paleoseismologic studies along the left lateral strike-slip Bogd fault, the
principal structure involved in the Gobi-Altay earthquake of December 4, 1957 (published magnitudes range
from 7.8-8.3). The Bogd fault is 260 km long and can be subdivided into 5 main geometric segments based
on the variation in strike direction. From west to east these segments trend: N095°E along West Ih
Bogd (WIB), N105°E along North Ih Bogd (NIB), N095°E along East Ih Bogd (EIB), N075°E
along West Baga Bogd (WBB), N115°E along North Baga Bogd (NBB), respectively.
Our goals are
to refine and complete slip rate estimates on the Bogd fault and to determine the ages and the slip
distribution of pre-1957 surface ruptures.Morphological analysis of offset streams, terrace risers, and ridges
at five sites (two at Ulaan Bulag on the WIB segment, one at Noyan Uul on the WIB segment, one at Bitut on
the NIB segment, and one at Zadgay Sayhr on the EIB segment) distributed across the three western
segments show that the slip along each segment is characteristic (the different offsets measured at a point
are multiple of the 1957 slip at this point). The horizontal components per event are between 5-5.5 m at
Ulaan Bulag, 4.4-5.2 m at Noyan Uul, 2.8-3.4 m at Bitut, and 4-5 at Zadgay Sayhr.
A paleoseismic study
on the western segment at Ulaan Bulag suggests a mean earthquake recurrence interval of ~ 3500-
3700 years for the late Pleistocene-Holocene. There have been four events during the past 14±1.65 ka
and three events since 10.6±0.71 ka (11.1±0.71 ka IRSL date), OSL dates bracket the age of the
penultimate event between 3.4±0.34 ka and 5.44±0.34 ka (5.05±0.3 ka IRSL date). At Bitut
(NIB) alluvial surfaces that have preliminary age dates between 3.71±0.46 ka (OSL) and 4.5±0.8 ka
(in situ 10Be) are only offset by the 1957 event. This is consistent with the interval between the 1957 and
penultimate events at Ulaan Bulag. We suggest that both the WIB and NIB segments ruptured together
during the penultimate earthquake along this section of the Bogd fault.Our still-in-progress slip rate study
indicates that the horizontal slip rate along the Bogd fault decreases progressively from ~1.3 mm/yr to
~0.8 mm/yr between Ulaan Bulag (WIB segment) and Bitut (NIB segment). This decrease reflects the
progressive change of kinematics from pure left-lateral strike-slip faulting to left-lateral-reverse faulting
consistently with the change of strike from WIB segment to NIB segment.
T21B-1960
Neotectonics of the Lopukangri Fault System Using Remote-Sensing Observations
Several hypotheses have been suggested to explain the regional kinematic development of the north- trending Tibetan rifts and include: (1) gravitational collapse due to crustal thickening, or following removal of the mantle lithosphere, (2) oblique convergence and growth of the Himalayan arc, (3) eastward extrusion of central Tibet relative to southern Tibet, (4) basal shear tractions related to underthrusting India, and (5) far field motions related to rollback of the Pacific slab. In order to assess the mode of extension of the Tibetan Plateau, we document the structure, neotectonics, and geomorphology of the Lopukangri rift using ASTER imagery, data from the 90 m Shuttle Radar Topography Mission (SRTM), and Landsat7 imagery. The Lopukangri rift is located in southwestern Tibet. It is bounded by the Lopukangri fault system (LFS), a system of northwest-dipping oblique-normal faults comprising a ~4 km wide shear zone along the eastern margin. The zone is comprised of curved fault segments that step basinward and link to form a north-south elongated S-shaped margin fault system. The 150 km long LFS is associated with oblique-slip faulting and observations of right-laterally displaced terrace risers suggest it is presently active. The stream network derived from the SRTM DEM suggests faulting along the eastern margin of the LFS is more dominant and is consistent with oblique-normal slip, as indicated by deflected streams, offset terrace risers, and scarp morphology. The northernmost LFS consists of en echelon faults that splay into two populations: a system of right-stepping oblique-normal slip faults that bound the eastern rift flank and left-stepping oblique-normal slip faults that bound the western rift flank within a valley width of 25 km. The width of the rift valley narrows to 4.5 km in the central area, and widens to 10 km in the south, north of the Indus-Yalu suture zone (IYS). The southern LFS bends at the IYS zone and cuts it with a right-separation of 15 km. The style of deformation observed in the LFS indicates oblique extension is important, and suggests boundary conditions, such as structural style and fault geometry, should be considered in more detail for any model of rift development in southwestern Tibet. The structural style of the LFS suggests models calling upon collapse or mantle delamination favoring east-west extension are inconsistent with the first order observations of oblique faulting present along the active Lopukangri rift. Models invoking extension need to be evaluated in terms of the relationship between field observations, remote-sensing interpretations, and processes occurring at depth.
T21B-1961
Geoelectrical Imaging of a Major Active Fault and Implications for Seismic Hazard Assessment in the City of San Miguel Uspantan, Quiche, Guatemala
A geophysical survey was carried out in the city of San Miguel Uspantan, Quiche, Guatemala, located along the seismically active Polochic fault which forms, with the Motagua Fault, the plate boundary between the N- American and Caribbean plates. During its history, the city of San Miguel Uspantan and its close region had experienced severely damaging earthquakes induced by the Polochic fault. The earthquakes generated by the fault during the last two centuries, although of moderate intensity, produced important damages. The importance of the destructions and the fact that even earthquakes of small magnitude can cause such important destructions indicate a possible site effect. Within the city, very little is known about the affected ground and the location and character of the investigated active fault as its surface expression was removed by human activity. For this reason, 2D Electrical Resistivity Tomographies (ERT) have been carried out in order to contribute to the geological knowledge of the recent tectonic structures by identifying and locating the active fault that traverses the city and by characterizing the thickness, nature, and physical proprieties of the ground on which the city was constructed. Firstly, geomorphologic observations and ERT profiles showed that the city was constructed on a Quaternary deposit of 20 m thickness with thixotropic properties that can be responsible for an amplification of seismic waves during an earthquake. Secondly, ERT profiles confirmed the presence of the fault within the city limits. The fault is expressed as a pronounced, near-vertical anomaly characterized by a sharp lateral resistivity contrast that not only allows interpreting the location of the fault zone to within a few meters, but also confirms that the fault was active until recent time. Finally, this study has significant implications for seismic hazard analysis in this tectonically active and populated region. The results of this work can help plan urban development and diminish the population exposure to natural hazards.
T21B-1962
A Closer Look at Salt, Faults, and Gas in the Northwestern Gulf of Mexico with 2-D Multichannel Seismic Data
The sedimentary wedge of the northern Gulf of Mexico (GOM) is extensively deformed and faulted by salt tectonics. It is also an area of biogenic and thermogenic gas and gas hydrate accumulation. To understand the interplay of salt tectonics, gas, and sedimentary history, we studied a dense grid of 2D multichannel seismic data covering an area of 33,800 km2 in the northwestern GOM, with an average grid spacing of 3.3 km. Salt is abundant in the study area. The distribution varies throughout the region from large diapirs and massifs in the east, to small isolated salt bodies in the west. Salt migration affects local topography and local bathymetric highs are salt diapirs and lows are areas of salt withdrawal. Two different styles of faulting are found within the study area. In the West, faults are longer, linear and trend NE-SW. They are parallel to slope which implies some degree of gravitation sliding. In the central and eastern part of the study area, faults are localized on salt bodies, resulting from diapirism. Acoustic blank zones caused by gas in the sediment column are common. Areas of gas enriched sediment found near the surface are assumed to contain biogenic gas. Vertical acoustic wipeout features, (aka "gas chimneys",) vent deeper thermogenic gas. While there is a high abundance of gas, little evidence of bottom simulating reflectors (BSR), a widely accepted indicator of gas hydrate, has been found.
T21B-1963
Davis Strait and Ungava Fault Zone: First Results From a Recent Geophysical Survey
The Davis Strait is a bathymetric high that separates the southern Baffin Bay and the northern Labrador Sea. These basins are the result of Cretaceous and Paleogene rifting and seafloor spreading between the North American plate and Greenland. Being one of the main tectonic features of the Davis Strait, the Ungava Fault Zone is associated with transform motion related to a northward movement of Greenland relative to North America during rifting and seafloor spreading in Baffin Bay and Labrador Sea. The plate tectonic reconstruction of the Davis Strait contributes significantly to the understanding of the geodynamic history of the North-American - Greenland plates, not only the Davis Strait area but also the area of Lancaster Sound and Nares Strait, where it could shed light into the so-called Nares Strait Conflict. It is still under debate whether the spreading between Greenland and Baffin Island was compensated by sinistral transform motion along the proposed Wegener-Fault. Thus Nares Strait (trace of the Wegener Fault) and Lancaster Sound (failed arm rift) are relicts of this scenario. The lack of evidence for transform motion between Greenland and Ellesmere Island contradicts this model and provokes the conflict. As major compression along the Eurekan Fold Belt overprinted the proposed transform motion along the Wegener Fault, the Ungava Fault Zone in the Davis Strait could give the missing information for the plate tectonic reconstruction. The onshore-offshore geology and structural setting of Baffin Island is analysed by recent publications that show an essentially non-volcanic continental margin at Baffin Island that is interrupted by a volcanic-style margin around Cape Dyer. Intensive magmatic activity during the initial opening phase is indicated by widespread seaward-dipping-reflector sequences (SDRS) north of Cape Dyer. On the other hand, the structural setting of the Greenland margin side is unclear. The identification of the corresponding conjugate pattern on Greenland side provides fundamental information for the historic motion along the Ungava Fault Zone. The geophysical data of the DAVIS GATE cruise in 2008 provide new data to determine structure and type of conjugate margin segments of Greenland and Baffin Island. As part of the DAVIS GATE project, a set of multichannel seismic, refraction seismic, magnetic and gravity profiles across the Davis Strait was acquired with RV Maria S. Merian in 2008. In detail, three crossing refraction lines with up to 25 ocean-bottom seismographs, and 1500 nm multichannel seismic lines (3000 m streamer length, 240 channels and 50 litres airgun-array) build the framework of the DAVIS GATE project. This presentation illustrates first results from the multichannel seismic survey in addition with first results from magnetic and gravity profiling.
T21B-1964
Collapse of the Cretaceous Helvetiafjellet Formation due to tectonic activity at Kvalvågen, eastern Spitsbergen
A variety of features recording disturbance of Mid-Cretaceous sediments are exposed in coastal cliffs at Kvalvågen, east Spitsbergen. The most striking of these features are large displaced blocks of Helvetiafjellet Formation sandstone (ranging from 5 to 25 meters across) that were dropped down into underlying shale- dominated sediments along normal faults. In addition to the displaced blocks much of the sandstone unit is missing along a 2 km stretch of coastal exposure and must have been slipped out of the plane of exposure. Several hypotheses have been proposed to explain the style and cause of the Cretaceous collapse at Kvalvågen including delta front collapse (Nemec et al., 1988), landslides into a submarine canyon (Steel et al., 2001), and collapse related to magmatic activity (Midtkandal et al., 2007). New structural data and field observations show that the orientations and style of deformation are not entirely consistent with the previous hypotheses and are better explained as the direct result of tectonically produced topography (i.e., a fault scarp). The deformation at Kvalvågen is the result of west-side-down displacement along a north-striking fault that crops out at the southern end of the cliff exposure. Tectonic disturbance in the area began in Hauterivian time and was over by the early Aptian. These outcrops are the only evidence of tectonic activity in the area during the Mid-Cretaceous and may be the result of displacement along a previously unrecognized extension of the Lomfjorden fault zone or related to regional stresses imposed by extensive sill intrusions during the formation of the High Arctic Large Igneous Province.
T21B-1965
Timing of Evaporite Deformation and Diagenesis Along the Billefjorden Fault Zone of Spitsbergen
Gypsum and anhydrite of the Middle Carboniferous Ebbadalen and Minkinfjellet Formations were deposited in an asymmetric rift basin, the Billefjorden Trough, in Spitsbergen, Norway. The Billefjorden Fault Zone (BFZ), the western boundary of the basin, had a long polyphase history including disputed Tertiary components. It is a basement seated structure attributed to Caledonian and possibly earlier deformation that was followed by reverse fault reactivation in the Devonian and major rift development in the Carboniferous. The Petuniabukta syncline is a large basin axial feature with 1 km of structural relief, which may be of Carboniferous, Tertiary or mixed age. The Mid-Carboniferous evaporites played an integral role in accommodating and recording deformation in the current study area. Foliations, fold axes, and elongation lineations in Middle Carboniferous sediments adjacent to basin bounding faults (both the Billefjorden fault and a sub-basin defining fault, the Lovehovden lineament) provide evidence as to kinematics. Strong foliations consistently oblique to bedding indicate dominant dip-slip motion with minor strike-slip components that vary along strike in a relay zone between two major faults of the BFZ. The orientations of deformation fabrics are not axial planar to the reported Tertiary Petuniabukta Syncline, but instead are associated with bedding-parallel simple shear associated with normal faults. Timing of deformational features is better constrained along the eastern Lovehovden Fault, where overlying strata and stratigraphic relationships across the fault constrain the normal slip as Carboniferous in age. Incomplete, post-consolidation diagenesis of gypsum is seen as pseudomorphic anhydrite replacing and overprinting the deformed gypsum fabrics. Anhydrite growth occurred on a prograde path, prior to maximum burial during Early Cretaceous. Late phase, secondary gypsum formed during exhumation and is concentrated along carbonate material, shear zones and cleavage planes of anhydrite grains. Variations in pore-fluid salinities and hydrodynamics of the basin are thought to explain local prograde and retrograde diagenetic variations, allowing for incomplete overprinting and the preservation of earlier textures. Deformed gypsum fabrics manifest in thin section as undulose extinction, subgrain development, polygonal grain geometries associated with dynamic recrystallization, and selenite porphyroclasts. Associated carbonate material is brittlely deformed. These results help constrain the timing and kinematics of reactivation along the BFZ.
T21B-1966
3-D Stress Variation at Ridge-Transform Systems From Bathymetry Variations and Plate Driving
Focal mechanisms of earthquakes at oceanic spreading ridge-transform plate boundary regions indicate that the 3-D stress tensor changes orientation significantly over a very short spatial scale at these locations. We investigate the role that 3-D elastic stress from spatial variations in bathymetry load plays in balancing the horizontal ridge-push tectonic stress, and we explore the possibility that this stress balance may describe the observed stress field indicated by fault orientations, focal mechanisms, and pattern of seismicity. Stress from bathymetry is computed semi-analytically, accounting for the loads of both topography at the surface and deflected moho topography, on a plate of uniform mechanical thickness but with varying flexural rigidity. We focus on areas where bathymetry is mostly constrained by multibeam soundings, and we adjust flexural parameters so that predicted gravity matches a new 1-minute resolution global gravity field. A variety of models are considered to represent tectonic ridge-push stress, including regionally uniform extension aligned perpendicular to the ridge axis and a spatially varying swell push stress derived from lithospheric contribution to the geoid. Considering the balance between topographic and tectonic sources of stress can also yield bounds on the absolute magnitude of the ridge-push component of plate driving stress.
T21B-1967
Normal Fault Growth on Mars
Displacement versus length relationships of faults on Earth and Mars have been used to describe and interpret the evolution of faults and fault systems, infer differences in the relative strengths of strata, and evaluate variations in fault-system response to differences in gravity from planet to planet. In this presentation, we focus on maximum throw versus trace length (Dmax/L) of continuously mappable faults and Dmax/L of individual fault segments. Fault analyses on Mars have the advantage of a planetary surface devoid of vegetation and largely unaffected by weathering and erosion. Areas on the flanks of Alba Patera, Mars, were chosen because they are well imaged by all generations of data coverage, contain fault systems that have a range of developmental characteristics, and formed in a relatively simple tectonic setting dominated by extension. Footwall and hanging wall cutoff traces of more than 300 faults were interpreted using Viking imagery and ArcGIS software. Throw was obtained by calculating the elevation difference between adjacent footwall and hanging wall points using Mars Orbiter Laser Altimeter data. Throw versus along-strike trace length plots were constructed for each interpreted fault. Single fault segments are defined as having one well-defined displacement maximum bounded by two near-zero displacement minima. Segments within a multi-segment fault were identified by counting displacement maxima along the fault trace. The number of segments incorporated into multi-segment faults is positively correlated with the fault trace length. In a plot of Dmax versus L, whole faults are distributed approximately along a locus of Dmax = K × Ln, where K = 5 × 10-4 to 5 × 10-2 and n = 1. This is in agreement with previous studies of faults on Mars. Single fault segments form a distinct population whose distribution is described approximately by the same equation but where K = 1.7 × 10-3. Dmax/L ratios for multi-segment faults represent an apparently self-similar evolution (n = 1) resulting from linkage and post- linkage development of single fault segments. In contrast, the distribution of single fault segments indicates that they do not grow by self-similar increments but evolve to higher Dmax/L ratios with successive slip events, a conclusion in accord with recent studies of terrestrial fault systems and coseismic surface ruptures.
T21B-1968
A complementary study of the Surprise Valley fault using a high-resolution shallow seismic reflection profile
Despite a lack of large historical earthquakes, the Surprise Valley fault system in northeastern California is presumed to be seismically active based on the presence of numerous Holocene fault scarps and a fault- controlled geothermal system. The Surprise Valley fault is believed to be similar to historically active basin and range faults such as the Dixie Valley fault in central Nevada. A 5-m-deep paleoseismic trench across the main fault revealed a steeply-dipping (~68 degrees) normal fault, while a vibrator reflection profile located a few kilometers north of the trench imaged a much more shallowly dipping normal fault (~25-30 degrees) at 0.5-1.0 km depth. It is unclear from these data if the steeply dipping fault exposed in the trench soles into the more shallowly dipping fault at depth, or if the currently active fault cuts the fault imaged in the reflection profile. In June of 2008, we conducted a 200-m-long high-resolution shallow reflection profile across the fault at the location of the paleoseismic trench in order to further constrain the geometry of the currently active range-front fault. The survey was conducted with a 48-channel seismograph, a 16-lb sledge, and 48 groups of six 100-Hz phones each, spaced at 2-m intervals. Optimization of P-wave arrival times for a tomographic image of the upper 50 meters shows velocities ranging from 1000 to 2500 m/s. Preliminary reflection processing, together with the optimized velocity section, suggests that the fault dips at an angle of 60 degrees in the upper 40 meters, and may suggest an antithetic fault in the hanging wall. The unprocessed data display strong reflections at depths to 200 meters. A detailed understanding of the dip angle of the Surprise Valley fault will impact the ongoing geothermal development and the assessment of seismic hazard in the area, both of which will vary based on the angle of the active fault system.
T21B-1969
The Faults and Hot Springs of Surprise Valley: Perspectives from Detailed Gravity and Magnetic Data
The central section of the Surprise Valley, NE California, in the westernmost section of the Basin and Range province contains a geothermal system related to extensional tectonics. Most geophysical research related to this geothermal system has focused primarily on the Lake City hot springs. Our work includes several other hot springs within the valley and their associated faults. We have used magnetic shallow crustal profiling in conjunction with gravimetric measurements to provide greater geophysical and geological understanding of the structures controlling the geothermal system. Potential fields data were collected along linear transects crossing inferred structural features using a Cesium Magnetometer, and Lacoste and Romberg and Scintrex CG5 gravity meters. These transects reveal two distinct fault types: intra-basin normal faults with significant vertical offset, and more complex faults representative of those previously mapped in the Lake City Fault Zone. The Lake City Fault Zone displays high frequency, high amplitude anomalies but does not appear to have considerable offset unlike faults trending parallel to the valley that have significant vertical offset. Our work provides additional detail that complements existing aeromagnetic and regional gravity data, suggesting that north-south trending normal faulting is occurring within the basin while the Lake City Fault Zone interacts obliquely with these faults in a NW-SE-trending direction. This research has better defined the locations of faults, and extended portions of the Lake City Fault Zone and surrounding faults. This bears significance for the prospective development of geothermal energy as a regional resource.
T21B-1970
Structural and Sedimentological Development of Pahrump Basin, Southern Nevada with Implications for Seismic Hazards
The purpose of this study is to (1) document potentially active faults and estimate possible earthquake magnitudes, (2) document and analyze sedimentation in a basin controlled by strike-slip and oblique faults, and (3) consider implications on regional development using Pahrump Valley, southern Nevada as a case study. The 2.5 million people living within the region would be significantly impacted by a major earthquake generated in the Valley. With an ever increasing population, the need for evaluation of seismic risk is becoming more important for land use planning in southern Nevada. Using data analysis of well logs, geophysical measurements, surface data from air photos, maps, and field observations, it is possible to document the 3D architecture of the basin-fill sediments and basin structure through abrupt changes in sedimentary facies. 3D modeling of the lithology and depositional environments of shallow basin fill improves the understanding of fault location, type, offset, and surface rupture length. Pahrump Valley is flanked by two documented Neogene (Quaternary) fault systems. The west side is dominated by the Stateline Fault zone, which is a continuous NW-striking right-lateral strike-slip fault zone extending 200 km from Mesquite Valley to Amargosa Valley, Nevada. In Pahrump Valley no fault scarp is present. The eastern side is bordered by the West Spring Mountains fault, which is a N-striking W-dipping normal fault with a right-lateral oblique component. The fault has a large scarp which is visible along the northeastern and east-central Valley border and smaller discontinuous scarps in the south. The 11-km-long central segment contains scarps up to 9.4 m high. On the basis of scarp profiles, the youngest event is estimated to be Pleistocene or early Holocene in age with a maximum fault displacement estimated at 1.8-2.0 m, which suggests an event of M 6.9. 3D basin models, derived from well log lithology, depict the locations of fault surfaces by showing abrupt changes in units and unit offsets among multiple wells. The lithologic data augmented with surface investigations and seismic data show a depositional environment dominated by alluvial deposits of coarse material (gravel and sand) and playa sediments (clay and soils) that interfinger. The playa sediments in the west valley exhibit right-lateral displacement of 10 km consistent with the Stateline fault zone.
T21B-1971
Slip Rates and Scarp Profiles, Using Terrestrial LiDAR to Quantify Deformation on the Aberdeen Fault, South-Eastern California
High resolution topographic measurements form the basis for slip-rate studies along active fault systems. However, meter-scale features such as fault scarps or offset terrace risers are insufficiently resolved on available remotely-sensed digital elevation models (e.g., 10 m NED), rendering it necessary to make field- based topographic measurements. Terrestrial or tripod mounted LiDAR (T-LiDAR) is a recent technology that has proved useful for making these types of high resolution measurements. In this study we used T-LiDAR to constrain the morphology of a series of fault scarps offsetting an alluvial fan and the flank of a cinder cone along the Aberdeen fault in the eastern Owens Valley. This site lies on the western edge of the Eastern California Shear Zone (ECSZ), a region defined by northwest striking right lateral faulting in south eastern California and western Nevada that accommodates 20-25% of Pacific-North American relative plate motion. The Aberdeen fault is one of a series of northeast striking normal faults that are thought to transfer slip from west to east within the ECSZ. With the T-LiDAR unit, we collected a ~17 million point data set covering an area of ~350 x 150 m from which we generated a 20cm DEM. These data allowed us to study and measure in great detail the series of fault scarps that offset the flank of the cinder cone and the alluvial fan. Located at the northeast end of the Aberdeen fault, this series of scarps exhibits a ~70° change in azimuth moving east along strike as the fault curves into the range front. Using a hillshade image created from the 20cm DEM, we selected three points from one continuous scarp that offsets the fan and the cinder cone from which we calculated a fault dip of 39°. The high resolution of the hillshade DEM also proved critical in interpreting slip history on this section of the Aberdeen fault by showing definitive evidence for two faulting events. Using a series of scarp profiles derived from the T-LiDAR data, we were able to measure geometric parameters of the scarps including the vertical offset on the alluvial fan (7.14 m). We interpret the fan to have formed after emplacement of the cinder cone (c. 90 ka), providing a maximum age constraint yielding a minimum dip-slip rate on the Aberdeen fault over the last 90ka of .13 ± 0.03 mm/yr. Morphologic diffusion analysis of topographic profiles extracted from the T-LiDAR data provide an alternative constraint on the age of the fault scarp. Preliminary results indicate a scarp age of ~21ka, resulting in a slip-rate that is ~4-times higher than the rate constrained by the cinder cone age. Further collection and analysis of T-LiDAR-derived data will more tightly resolve fault geometry and event chronology at this site and yield further insight into the role of the Aberdeen fault in ECSZ tectonics.
T21B-1972
Seven Post-Pennsylvanian Structural Events in the Dry Hills and Northern Osgood Mountains: Evidence of Late Paleozoic Tectonism and New Getchell Fault Offset Data
An incomplete understanding of the number and style of deformational events in north-central Nevada complicate economic and stratigraphic studies. To help unravel this structural history, we mapped and measured structures in the Dry Hills, northern Osgood Mountains, Humboldt County, Nevada. This research identified at least seven Pennsylvanian age or younger deformational events. The location was selected because it contains the Roberts Mountain allochthon (RMA), the Pennsylvanian (IP) to Permian (P) age Etchart Formation that was deposited atop the RMA, and the Golconda allochthon (GA). Measured folds and faults provide evidence of at least four deformational events that occurred between emplacement of the RMA and GA. Evidence for Late Paleozoic deformational events was recently well documented in other regional locations, but this location is farther north than previous studies. This research is also important because it provides information about the style and regional extent of these tectonic events. Structural mapping and data analysis reveal that the Etchart Formation contains four upright fold sets (SE-, NNW-, SSW-, NE-trending, in chronologic order) and three fault sets (N-striking normal, ENE-striking thrust, NNE-striking thrust), which may relate to the three youngest fold sets. The relative ages of the folds and faults are constrained via cross-cutting relationships, overprinting, and the oldest fold set being confined solely to the lowest stratigraphic unit within the Etchart Formation. These structures developed prior to and during emplacement of the GA. Two angular unconformities, the C5 and P1 angular unconformities of Trexler et al. (2003), are newly recognized in the area. The SW-trending fold set underlies the C5 unconformity. The P1 unconformity is angular indicating a deformational event and has been confirmed with fusulinid-based ages. The P1 and C5 unconformities correlate to other locations in northern Nevada implying regional scale late Paleozoic tectonic events. IP and P fold geometries and stratigraphy differ greatly from previously studied nearby areas, implying a different deformational history here than in other study areas. Post-Etchart deformation includes the emplacement of the GA and movement on the Getchell fault. Between the northern Osgood Mountains and the Dry Hills, the frontal thrust of the GA is offset by the Getchell fault. The Getchell fault is known for structurally controlling local ore-bearing units. Offsets of units and older faults indicate that the Getchell fault is a right-oblique normal fault. The offset also allows a new estimation of the minimum right-lateral offset on the Getchell fault of approximately 1.8 km since emplacement of the GA.
T21B-1973
New Constraints on Late Pleistocene - Holocene Slip Rates and Seismic Behavior Along the Panamint Valley Fault Zone, Eastern California
Space-time patterns of seismic strain release along active fault systems can provide insight into the geodynamics of deforming lithosphere. Along the eastern California shear zone, fault systems south of the Garlock fault appear to have experienced an ongoing pulse of seismic activity over the past ca. 1 kyr (Rockwell et al., 2000). Recently, this cluster of seismicity has been implicated as both cause and consequence of the oft-cited discrepancy between geodetic velocities and geologic slip rates in this region (Dolan et al., 2007; Oskin et al., 2008). Whether other faults within the shear zone exhibit similar behavior remains uncertain. Here we report the preliminary results of new investigations of slip rates and seismic history along the Panamint Valley fault zone (PVFZ). The PVFZ is characterized by dextral, oblique-normal displacement along a moderately to shallowly-dipping range front fault. Previous workers (Zhang et al., 1990) identified a relatively recent surface rupture confined to a ~25 km segment of the southern fault zone and associated with dextral displacements of ~3 m. Our mapping reveals that youthful scarps ranging from 2-4 m in height are distributed along the central portion of the fault zone for at least 50 km. North of Ballarat, a releasing jog in the fault zone forms a 2-3 km long embayment. Displacement of debris-flow levees and channels along NE-striking faults that confirm that displacement is nearly dip-slip, consistent with an overall transport direction toward ~340°, and affording an opportunity to constrain fault displacement directly from the vertical offset of alluvial surfaces of varying age. At the mouth of Happy Canyon, the frontal fault strand displaces a fresh debris-flow by ~3-4 m; soil development atop the debris-flow surface is incipient to negligible. Radiocarbon ages from logs embedded in the flow matrix constrain the timing of the most recent event to younger than ~ 600 cal yr BP. Older alluvial surfaces, such as that buried by the debris-flow lobe, exhibit progressively larger displacement (up to 10-12 m). Well-preserved bar and swale morphology, incipient varnishing of surface boulders, and weak soil development all suggest that this surface is Late Holocene in age. We are working to confirm this inference, but if correct, it suggests that this fault system may have experienced ~3-4 events in the relatively recent past. Finally, preliminary surface ages from even older surfaces along this portion of the fault zone place limits on the slip rate over Late Pleistocene time. Cosmogenic 10Be surface clast dating of an alluvial surface with well-developed pavement and moderate soil development near Happy Canyon suggests a surface age of 30-35 kyr. We are working to refine this estimate with new dating and soil characterization, but our preliminary reconstructions of displacement of this surface across the two primary fault strands are consistent with slip rates that exceed ~3 mm/yr. Overall, these results are consistent with the inference that the Panamint Valley fault zone is the primary structure that accomplishes transfer of right-lateral shear across the Garlock Fault.
T21B-1974
Geomorphic mapping of the southern Maacama fault based on LiDAR data
The Maacama fault is an active strike slip fault, and a potentially significant seismic source, within the San Andreas transform system. The fault is located east of and parallel to the San Andreas fault in Sonoma and Mendocino counties, California and is divided into a northern and southern section based on a NW to NNW change in strike. The southern segment comprises 54 km of the fault's 144 km total length and is primarily located in an upland area traversing mountainous terrain. Strain is thought to transfer northward from the East Bay fault zone along the Rodgers Creek fault and, through a right step, to the Maacama fault. LiDAR data collected in a 1-km-wide swath along the southern Maacama fault, as part of the GeoEarthscope project, were used to produce a bare-earth digital elevation model, from which hillshade, topographic contour, slope, and curvature maps with 0.5- to 1-m-resolution were derived. Mapping was primarily conducted digitally in a GIS environment, and interpretation of LiDAR data was supplemented with aerial photograph interpretation and field inspection. Primary, Holocene-age fault-related geomorphic features, consisting of scarps and dextrally offset drainages, define the southern Maacama. These features are sparsely distributed and comprise less than 20% of the fault length. The fault scarps define a sequence of left-stepping, en echelon fault segments with an average segment length of 230 m. By contrast, the northern Maacama fault is better defined geomorphically. The poor expression of the southern Maacama is likely due to the presence of active hillslope processes and low levels of seismicity. Seismicity along the southern segment is lower than that of the northern segment. The Coast Range uplands, primarily composed of Franciscan Complex, is characterized by numerous landslides and experiences annual precipitation of 75 to 180 cm. There is approximately 30 km of overlap between the northern end of the Rodgers Creek fault and the southern extent of the Maacama fault, and in this region the Rodgers Creek fault is better expressed geomorphically. Because climate and lithology are similar across this region, we attribute the stark differences in geomorphic appearance to tectonic activity. This difference suggests that over Holocene and late Pleistocene timescales, strain has been preferentially accommodated along the Rodgers Creek fault and is gradually transferred northward to the Maacama fault.
T21B-1975
A Study of the San Andreas Slip Rate on the San Francisco Peninsula, California
The most recent large earthquake on the San Andreas Fault (SAF) along the San Francisco Peninsula was the great San Francisco earthquake of April 18, 1906, when a Mw= 7.8 event ruptured 435-470 km of the northern SAF. The slip rate for this segment of the SAF is incompletely known but is important for clarifying seismic hazard in this highly urbanized region. A previous study south of our site has found an average slip rate of 17±4 mm/yr for the late Holocene on the San Francisco Peninsula segment of the SAF. North of the Golden Gate, the SAF joins the San Gregorio Fault with an estimated slip rate of 6 mm/yr. A trench study north of where the two faults join has produced an average late Holocene slip rate of 24±3 mm/yr. To refine slip-rate estimates for the peninsula segment of the SAF, we excavated a trench across the fault where we located an abandoned channel between the San Andreas and Lower Crystal Springs reservoirs. This abandoned channel marks the time when a new channel cut across the SAF; the new channel has since been offset in a right-lateral sense about 20 m. The measured amount of offset and the age of the youngest fluvial sediments in the abandoned channel will yield a slip rate for the San Francisco Peninsula segment of the SAF. We excavated a trench across the abandoned channel and logged the exposed sediments. Our investigation revealed channel-fill alluvium incised and filled by probable debris flow sediments, and a wide fault zone in bedrock, west of the channel deposits. The most prominent fault is probably the strand that moved in 1906. We completed a total-station survey to more precisely measure the offset stream, and to confirm that the fault exposed in the trench aligns with a fence that is known to have been offset 2.8m during the 1906 earthquake. We interpret the debris flow sediments to represent the last phase of deposition prior to abandonment of the old channel. We collected samples for radiocarbon dating, optically stimulated luminescence (OSL) dating, and for detailed sedimentary descriptions. Once we obtain an age for the debris-flow sediments, which closely pre-date the time of channel abandonment, we expect to be able to calculate a maximum slip rate that will contribute to a better understanding of the SAF on the San Francisco Peninsula.
T21B-1976
Slip History and Evolution of the Hat Creek Fault, Northern California
Normal faults commonly exhibit unique surface features in basalt such as vertical scarps and fault-trace monoclines that provide clues to the fault evolution. The Hat Creek fault, 25 km north of Lassen volcano in northern California, is a segmented fault system within Pleistocene and younger basalts. The fault is located along the western boundary of the Modoc plateau in the extended backarc of the Cascades. The fault geometry tells of a varied extensional history that likely reflects a complex interplay between tectonic and magmatic influences. In response, the northern portion of the fault system migrated progressively westward, abandoning older scarps in its wake, whereas the southern portion continues to utilize Pleistocene slip surfaces. This spatial evolution has created three distinct scarps. From oldest (easternmost) to youngest (westernmost), they are informally identified as: the rim (max. throw of 352 m), the pali (max. throw of 174 m), and the active scarp (max. throw of 65 m). The rim is oriented N-S, consistent with the regional E-W extension direction, and consists of 7 predominantly right-stepping segments (NNW oriented) that are physically linked through lower ramp breaches. This geometry implies a clockwise rotation of the stress field after the segments developed, with linkage driven by right-lateral oblique motion. Throw profiles along the rim illustrate mechanical interactions and partitioning of displacement between adjacent segments. The pali is a relatively younger fault plane located up to 3.3 km west of the northern portion of the rim. The pali is oriented NW-SE and consists of 5 left-stepping segments that are physically connected through upper ramp breaches, also consistent with right-lateral oblique motion. The pali likely nucleated along its central segment, where throw is maximized, in response to a magmatic perturbation of the stress field (causing a local NE-SW extension), possibly related to dike injection that culminated in the creation of the nearby volcanic edifice, Cinder Butte (38 ± 7 ka). The association with Cinder Butte is most apparent along the northernmost segment of the pali, which curves around the edifice. Southward propagation of the pali eventually resulted in linkage with the rim, which is still active south of the linkage point. The development of the active scarp was prefaced by the eruption of the 24 ± 6 ka Hat Creek basalts (about 50 m thick) in the hanging wall valley of the Hat Creek fault. These lavas pooled against the pali in the north and the rim in the south. Post-eruption slip along the fault forced the upper fault tip vertically through the Hat Creek basalt. Vertical growth was incremental, presumably over multiple earthquake events, resulting in monoclinal folding of the surface above the buried fault tip. The fault eventually pierced the surface, breaching the monocline and forming the vertical active scarp. The active scarp consists of 7 left-stepping segments in various stages of linkage (mostly unlinked, but 2 with upper ramp breaches) that trace the pali in the north and the rim in the south. This geometry, combined with left-stepping fractures along unbreached portions of the monocline upper hinge, suggest a recent component of right-lateral motion where the active scarp traces the pali, consistent with the contemporary regional E-W extension. The monocline shows variable states of disaggregation that point to the ongoing effects of earthquake activity. Although there have been no historic events along the fault, the length of the active scarp suggests possible M6.5 events.