T11C-1266 0800h
The Seismicity of Stable Continental Regions
The large impact on humans of large magnitude earthquakes in stable continental regions provides a strong motivation to study the systematic nature of their occurrence, to improve the assessment of seismic hazards. Unfortunately, limited historical information of these events complicates studies to better understand these events. The existing main hypotheses to explain their occurrence are (1) reactivation of zones of weakness; (2) localization of stress by physical stress concentrators; (3) crustal weakening by fluids; (4) anomalous high temperatures; and (5) stress changes due to deglaciation or sediment loading. In this study stable continental seismicity patterns were compared with potential field data (Bouguer gravity and aeromagnetic anomaly maps are made). The study cannot disprove any of the existing hypotheses for the occurrence of these events, but it is possible to evaluate some of them. A significant portion of stable continental seismicity follows long linear geophysical trends (500-2000 km) that are correlated with major tectonic features, such as rifted margins and suture zones. The correlation of seismicity with localized stress concentrators is not consistent with this observation, which rather supports the zone of weakness hypothesis. However, in several cases of seismicity not observed in linear trends, the localized stress concentrator hypothesis is instead favorable. Rifted margins commonly have both sedimentary loads and high-density mafic lower crust, both of which provide a stress perturbation. The observed correlation with rifted margins is strong in Australia but less pronounced for eastern North America and is therefore not uniform. The correlation of seismicity with geophysical anomaly maps is an effective means of testing the multiple hypotheses that have been advanced for the genesis of earthquakes in Stable Continental Regions. We favor the view that multiple hypotheses are valid for stable continental earthquakes. Additional studies will help better our assessment of seismic hazards in Stable Continental Regions.
T11C-1267 0800h
Localized weak zone and their role in generation of intraplate seismicity: Effects of weak zone geometry and rheology
In intraplate seismic zones (e.g. the New Madrid Seismic Zone, NMSZ, in the south-central United States) the source of stress that drives earthquake generation is very complex. One hypothesis proposes that geologically transient, short-lived bursts of earthquakes lasting 10's of thousands of years may results from perturbations of the local or regional stress field. This causes relaxation of a lower crustal weak zone which may drive repeated earthquakes. The number of earthquakes occurring in a given time is dependent on the geometry and rheology of the weak zone. In this study, we investigate the relevant parameter space as it affects the concentration of stress at the base of the seismogenic fault. Using finite element techniques which employ contact surfaces to model discrete faulting events and a maximum shear stress criteria evaluated at each node, variable stress drops are also considered. We find that model parameters representing the geometry and rheology (Maxwell, standard linear solid, and power-law) of weak zone are particularly important. Other parameters that are considered include earthquake stress drop, background tectonic stress, and maximum shear stress at failure. Results show that solutions are non-unique. With the addition of existing geological evidence, however, we can place some constraints on range of parameters which satisfy observations from the NMSZ. Initially, 2D antiplane models have been run to efficiently identify trends in the parameter space. This information is then used to guide the generation of 3D models, which are numerically costly but geologically more realistic. Some of the trends found using the 2D models as follows. When higher power-law exponents are used, the frequency of earthquakes is initially very high and the weak zone reaches a fully relaxed state relatively quickly. Thus, higher power-law exponents produce higher energy dissipation rates. As weak zone width increases, the total number of earthquakes occurring before the weak zone is totally relaxed also increases. There is an increase in the total time required for full relaxation as well. Using power-law rheologies in 2D, we can have 10-12 earthquakes in weak zones as narrow as 3 km. 3D models using the same weak zone width yield 5-6 earthquakes in first 3000 years. The duration of this earthquake sequence can be expanded by increasing the weak zone width. Thus, findings in 2D can be used to limit the number of numerically expensive 3D models which must be run to successfully define a parameter space which satisfies observations from the NMSZ.
T11C-1268 0800h
Paleoseismological Evidence of Large Pre-historical Earthquakes on the Western Haiyuan Fault Near Tianzhu, Gansu Province
The Haiyuan fault is a major active left-lateral fault that bounds the Tibetan Plateau to the northeast. Two great earthquakes have ruptured the fault system in the past century: in 1920 (M=8.7) along the eastern Haiyuan fault and in 1927 (M=8-8.3) on a thrust fault system north of the western Haiyuan fault. Studying this active fault has profound implications on the framework of deformation of the Tibetan Plateau, and on the mechanics of continental deformation in general. Previous studies have mainly focused on the slip rate of the fault, which remains debated. Paleoseismological investigations on the fault are sparse and mainly in the reach of the 1920 earthquake rupture, or along the eastern half of the Haiyuan fault. The seismic history of the 260-km-long western Haiyuan fault is largely unknown. Despite clear evidence in the field of Holocene activity, this segment of the fault bears no trace of a large earthquake in the past several centuries. It has therefore been reported as a major seismic gap (the `Tianzhu gap') with high potential seismic hazard. We have opened two trenches in a small pull-apart basin along the fault near longitude 103°30', north of Songshan in the Gansu province. The trenches are about 3 m deep, 4-5 m wide, and 20-30 m long. The excavation exposes sedimentary layers of alternating colors: black and yellowish white. The black layers are composed of organic-rich silty to clayey deposit, and yellowish white layers of coarser-grained sandy deposit. The main fault zone is readily recognizable in the exposures, by the disruption and tilting of the layers. The stratigraphy shows no obvious hiatus in deposition. 5-6 paleoseismic events are identified in the 2-3 m deposit below the ground surface. Charcoal is abundant in exposures and has been sampled. The initial batch of 10 samples has been selected for C14 dating, currently in process. The dates will help us to quantify the frequency of large ground-breaking earthquakes. If combined with the amounts of slip-per-event of these events, it will shed light on the slip rate in the last several millenniums.
T11C-1269 0800h
Joint Inversion Of Local And Teleseismic Data For The Crust And Mantle Structure Of The Chinese Capital Region
The Chinese Capital (Beijing) region is located in the intersection of the Yanshan and Taihangshan uplifts in North China. It is one of the regions with the strongest continental earthquakes in the world such as the 1976 Tangshan earthquake (M 7.8) which killed 240,000 people. Hence the determination of the crust and mantle structure of this region is very important for understanding the regional tectonics and for the reduction of earthquake hazards. Since October 2001 a new digital seismic network with 107 stations has been installed in this region, which is the most advanced and densest regional digital seismic network in mainland China. In this study we used 48750 P-wave arrival times from 2973 local events and 12249 travel time residuals from 234 teleseismic events recorded by this new digital seismic network. We adopted the local and teleseismic joint inversion approach by Zhao et al. [1994] and obtained a high-resolution three-dimensional (3-D) P-wave velocity model of the crust and mantle down to a depth of 1000 km. The resolution is 50 km in the horizontal direction, and in depth it is 4-17 km in the crust and 30-50 km in the mantle. The complex morphology of the Conrad and Moho discontinuities was taken into account in the tomographic inversions. Our 3-D velocity model provides new insights into the geological structure and tectonics of this region. The velocity images of the upper crust reflect well the surface geological, topographic and lithological features. In the North China Basin, the depression and uplift areas are imaged as slow and fast velocity belts oriented in NE-SW direction. The trend of velocity anomalies is the same as that of major faults and tectonics. Paleozoic strata and Pre-Cambrian basement rocks outcrop widely in the Taihangshan and Yanshan uplift areas, which exhibit strong and broad high-velocity(high-V) anomalies in our tomographic images, while the Quaternary intermountain basins show up as small low-velocity(low-V) anomalies. Our tomographic images of the deep crust and upper mantle layers also show different velocity features under different tectonic units. In the Yanshan uplift, a high-V patch is visible from the crust to about 300 km depth, which may reflect that the Yanshan uplift is a stable block without strong earthquakes. In the North China Basin, especially around Beijing, Tianjin and Tangshan,broad low-V anomalies are visible in the lower crust and at depths of 50ÍŽ100 km. Beneath the Bohai Bay, low-V anomalies exist and extend down to about 120 km depth. These results suggest that the lithosphere becomes thinner in this region. From the Taihangshan uplift to the Shanxi rift low-V anomalies exist from the uppermost mantle to about 300 km depth. In this region, lava rocks are distributed broadly and the Datong volcano is located. In addition, most of the large earthquakes, such as the 1976 Tangshan earthquake and the 1679 Sanhe earthquake (M 8.0), generally occurred in high-V areas in the upper to middle crust. However, in the lower crust and uppermost mantle under the source zones of the large earthquakes, low-V and high-conductivity anomalies exist. This result suggests the occurrence of strong earthquakes is related to the deep structure and processes in the deep crust and upper mantle.
T11C-1270 0800h
Distribution Pattern of Terrestrial Heat Flow in Bohai Bay Basin, North China
New temperature data from wells in Bohai bay basin increasing associated with the enhancement of oil and gas exploration there provides more reliable information about studying on Terrestrial heat flow pattern. Based on the data from 88 systematic continuous temperature logging curves and more than 1000 well test temperature data, along with the corresponding thermo-physical parameters of rock samples, here we determined 53 heat flow data and estimated other 172 according to thermal resistance method, then the distribution Pattern of heat flow in Bohai Bay basin is presented. Heat flow in Bohai bay basin is relatively large than those in the surrounding mountain areas. For instance, heat flow of Yanshan, north of the basin, is only low as 25 ~ 54 mW/m2, and less than 50 mW/m2 for Taihang mountain to the west, the average heat flow of Luxi Uplift is about 54 mW/m2. Crustal thickness of regions outside the basin to the west and north approximating to 36~44km, apparently is larger than that of basin, which maybe accounts for the high heat flow in Bohai bay basin. Those regions of relatively thin crust within the basin are of middle-high heat flow. Heat flow in such depressions as the Lower Liaohe, Bozhong, Jiyang and Yongqing area northeast of Jizhong Depression, together with Bohai offshore, for example, are all larger than 64 mW/m2, and even high as 70 mW/m2 for some regions with mantle upwelling. Low heat flow appears in those areas with relatively thick crust. For instance, heat flow in Linqing Depression, southwest margin of Jizhong Depression and southern Huanghua Depression, are all less than 64 mW/m2, even less than 60 mW/m2 for those areas with mantle downwelling. Heat flow pattern in Bohai Bay basin is negative correlation with crustal thickness, for those regions with relatively crustal thinning, heat derived from the deep earth is more due to the large lithospheric extension, resulting in the high heat flow; while for those with crustal thickening, heat flow is relatively low owing to small heat from earth interior. Mantle upwelling and crust thinning associated with episodic lithospheric extension in Bohai Bay basin formation, results in deep magma's intrusion into shallow crust and/or eruption into surface, which accounts for the high heat flow background of basin. Bohai Bay Basin is now continuing to subside, along with the heat attenuation process due to not achieving the heat balance.
T11C-1271 0800h
Anatexis in Himalayan Crust: Evidences and Implications
Here we present results of chemical features of migmatites in Higher Himalayan Crystallines (HHC) and K-Ar and Ar-Ar ages of the leucosomes (type-I and type-II) in the migmatites. The mass leucosomes distributed in HHC were generated by partial melting and consequently they are ideal specimens to determine the timing of anatexis, to study the relationship between migmatization and leucogranite formation, and to understand the role of partial melting in crustal evolution in continental collision orogenic belt. Mass balance relationship among mesosome, type-I leucosome and melanosome implies that they are likely generated in the same process. The compositions of Type-I leucosome are identical with those of melt produced by dehydration melting of biotite-plagioclase gneiss but different from those of type-II leucosome which compositions are similar to HHL. These compositional characteristics of leucosomes reflect that Type-I leucosome is the product of crystallization from melt generated by partial melting of mesosome in the source region, but Type-II leucosome and HHL probably underwent crystallization differentiation of plagioclase during melt aggregation and migration. Prime partial melting occurred at 22.7-24.7 Ma based on the date of type-I leucosome. The ages of type-II leucosomes (ranging from 14.82 to 18.37 Ma) are consistent with that of HHL. Very young age of 6.2-8.3 Ma of type-II leucosome provides a new time constraint on magma activity in the central segment of Higher Himalayas. These ages reveal some dynamic correlation summarized as follows. (1) The age of about 24 Ma (the beginning of anatexis), which is a little younger than that of MCT movement, does not support the view that anatexis was derived from MCT movement; on the contrary, but rather, the anatexis may have played an important role in the formation of MCT and Southern Tibetan Detachment System. (2) Decompression melting associated with STDS movement and crust uplifting contributed greatly to the increase of melt fraction and consequently formation of massive leucogranites. (3) Wide crust melting provides a strong positive feedback on crust extension and uplift further. We propose that the "feedback" mechanism between anatexis and crustal movement is a basic way for crustal evolution during post-continental-continental collision.
T11C-1272 0800h
Sr, Nd, Pb and Hf Isotopic Compositions of Late Cenozoic Alkali Basalts in South Korea: Evidence for Mixing Between the Two Dominant Asthenospheric Mantle Domains beneath East Asia
We determined the Sr, Nd, Pb and Hf isotopic compositions of late Cenozoic basaltic rocks from six lava-field provinces in South Korea, including Baengnyeong Island, Jogokni, Ganseong area, Jeju Island, Ulleung Island and Dog Island, in order to understand the nature of the mantle source. The basalts have OIB-like trace element abundance patterns, and also contain mantle-derived xenoliths. Available isotope data of late Cenozoic basalts from East Asia, along with ours, show that the mantle source has a DMM-EM1 array for northeast China and a DMM-EM2 array for Southeast Asia. We note that the basalts falling on an array between DMM and an intermediate end member between EM1 and EM2, are located between the two large-scale isotopic provinces, i.e., around the eastern part of South Korea. The most intriguing observation on the isotopic correlation diagrams is spatial variation from predominantly EM2 signatures in the basaltic lavas toward increasingly important addition of EM1, starting from Jeju Island to Ulleung and Dog Islands to Ganseong area, and to Baengnyeong Island. This is without any corresponding changes in the basement and the lithospheric mantle beneath the region. These observations suggest that the asthenospheric mantle source is dominant for the Cenozoic intraplate volcanism in East Asia, which is characterized by two distinct, large-scale domains. Previous studies on East Asian Cenozoic volcanic rocks have invoked origins by either plume activity or decompressional melting in a rift environment. On the basis of our new trace element and isotopic compositions which have OIB-like characteristics, we prefer a plume origin for these lavas. However, because tomographic images do not show distinct thermal anomaly that would be interpreted as a plume, we suggest that the magmatism might be the product of small, difficult to image multiple plumes that tapped the shallow part of the asthenosphere (probably the transition zone in the upper mantle).
T11C-1273 0800h
Metamorphic Evolution In The Northern Kontum Massif, Central Vietnam
Relict kyanite-bering rocks were found from the sillimanite zone of Kham Duc formation in the northern Kontum massif, central Vietnam. Kyanites occur as partially replaced by muscovite, margarite and quartz in garnet-muscovite (Grt-Mus) schist and included or surrounded by plagioclase in garnet-biotite-sillimanite (Grt-Bt-Sil) gneisses. Three metamorphic stages (M1-M3) are distinguished based on mineral paragenesis and compositions of these rocks. M1 stage is characterized by the presence of kyanite, high grossular (Xgrs = 0.21) and low pyrope (Xprp = 0.04) garnet and low anorthite content (Xan = 0.02-0.04) plagioclase. M2 stage is characterized by low grossular (Xgrs = 0.08) and high pyrope (Xprp = 0.14-0.18) garnet and high anorthite content (Xan = 0.36-0.45) plagioclase and appearance of sillimanite in gneisses. M3 stage is recorded in the rim of the garnet porphyroblast and resorbed garnet with coronal plagioclase. Pressure-temperature (P-T) conditions of M1-M3 stages calculated using geothermobarometers are M1: 11.8 kbar ($\sim$42 km) and 464 $\deg$C; M2: 6.8-7.0 kbar ($\sim$25 km) and 711-722 $\deg$C; M3: 4.0-4.6 kbar ($\sim$16 km) and 523-596$\deg$C. The P-T path of M1 to M2 indicates large temperature increase ($\sim$250$\deg$C) and pressure decrease (4.8 kbar). Although age and tectonic study are waited for detail studies, the preliminary result of 17 km exhumation with temperature increase may suggest significant lithospheric extension accompanying asthenospheric upwelling.
T11C-1274 0800h
Geomorphology of the Newly Identified Jid Strike-Slip Fault in the Altay Mountains, Western Mongolia
A right-lateral strike-slip fault (herein named the Jid fault) runs north-south for 60 km in the eastern Kharkhiraa Range of the Mongolian Altay. The Altay Mountains contain seismically active thrust and strike-slip faults that have generated numerous recorded earthquakes of magnitude $>$ 7. Identifying active structures such as the Jid fault has implications for estimates of seismic hazard and for understanding the regional tectonics. The fault shows indications of recent movements at scales ranging from drainage re-organisation at the kilometre scale, to ground deformations likely to result from movements during a single earthquake. From the scale of the ruptures and the observed length of roughly 60 km the most recent earthquake may have had a magnitude of about 7.5. In addition, numerous horizontal offsets of about 10-15 m, and vertical scarps of $<$ 5 m in the youngest alluvial deposits show the likely cumulative Holocene movement. The fault has many excellent examples of features associated with oblique-slip active faults, including stream displacements, pull-apart basins, shutter ridges, and river piracy. Along most of the Jid fault trace there is a slight vertical component of motion. The vertical component is introduced by bending along the fault and is normal in some places, and reverse in others. Parts of the fault show structural complexity at the surface with numerous parallel strands having both normal and reverse components of motion. The complexities probably form a `flower structure' in cross-section. We suspect from the geomorphology that the fault may have an overall normal conponent, which is not typical of deformation within the Altay. Assuming that the last period of alluvial fan deposition was at the last glacial maximum (approximately 10-15 ka ago), the horizontal offsets of 10-15 m would imply roughly 1 mm/yr of strike-slip motion. If the fault fails in magnitude 7.5 earthquakes with 3 m of slip in each, the average time between events will be about 3,000 years.
T11C-1275 0800h
Crustal velocity structure in the Changbaishan volcanic area, NE China, determined from a wide-angle refraction/reflection experiment and a temporary broadband array
The Changbaishan volcanic region, located along the border between China and North Korea, has had several large eruptions in the past few thousand years. However there has been little volcanic activity for several hundred years. So far studies of this active volcanic group have been limited and have not been widely published. In the summer of 1998, the State University of New York at Binghamton, in collaboration with the Research Center of Exploration Geophysics (RCEG) of the Chinese Seismological Bureau, installed 19 portable, broadband seismic stations on the Chinese side of the region of the Changbaishan. The broadband stations were borrowed from IRIS/PASSCAL, installed in June 1998, and were in operation for 3--8 months. During that summer, the RCEG conducted a wide-angle experiment. The active source experiment consisted of two primary lines and several shots and receivers distributed over the region. One of the primary lines trended north-south across the Tianchi volcano (the main volcanic edifice in the area), while the other trended east-west to the north of the Tianchi. We estimated crustal velocities by modeling receiver functions derived from seismograms collected at the broadband network as well as inversion of the travel-times from the wide-angle experiment. Both results resolved a low-velocity zone (LVZ) in the upper crust (below about 5--10 km depths) to the north of the Tianchi volcano and a 5--10 km thickening of the crust under the Tianchi. This LVZ is defined by sharp velocity gradients and velocities that are at most 0.8 km/sec below the surrounding crust. The LVZ is approximately 10--15 km thick and is likely 5--20$\times 10^3$ km$^2$. The nature of the LVZ can be explored in view of the previous geophysical and geochemical studies of this area.
T11C-1276 0800h
Three-dimensional Mechanical Modeling of the GPS Velocity Field around the Northeastern Tibet and Surrounding Region
The northeastern Tibet and surrounding region is an intense active tectonic area with great potential of strong earthquakes. Active tectonics is featured mainly by left-lateral transpressive deformation along the Qilian Shan and right-lateral transtensional deformation between the Alashan and the Ordos blocks out of the Tibet. This region gives a good example to investigate how the strain rate is partitioning around the plateau edge and how the remainder strain rate of Tibetan deformation is propagating towards the interior of the Eurasia plate with relative to the Indian-Eurasia convergence. In this study, we use three-dimensional mechanical models incorporated the main active faults as Coulomb-type friction zones to simulate the surface velocity field as dense GPS data shown. We simplify the rheological structure of lithosphere as a frictional upper crust underlying with the viscoelastic lower crust with available heat flow. Topographic loading on the model surface and hydrostatic pressure on the model base are added. The models are solved with Adeli finite element code, and modeled velocities are checked using the chi-square merit function with dense GPS data. We test the model with fault friction ranged from 0.4 to 0.01 on different major fault systems. Results show: (1) with a relatively high fault friction ~0.4-0.2, the northeastern Tibet and surrounding region could behave like diffusive deformation, but the modeled velocities and the strain rate seem unable to fit the GPS data and the seismic data, respectively. (2) To best fit the GPS data, a relatively low fault friction ~0.05-0.02 is needed. At this condition, the deformation is localized mainly around the Qilian Shan in northeastern Tibet and between the Alashan and the Ordos blocks out of the Tibet, consistent with the active tectonics and the strong earthquake distribution. (3) Changes of the fault friction out of the Tibet seem to affect the modeled velocity quite small, suggesting that the northeastern Tibetan boundary might be more sensitive to cause the present-day deformation as the dense GPS indicated. Fifteen model experiments give some first-order results. We conclude: (1) The northeastern Tibetan boundary may presently represent as a mechanically weakness zone at least in the upper crust relative to its surrounding regions. (2) This weakness boundary absorbs a significant amount of crust deformation relative to the Indian-Eurasia convergence from oblique thrusting on its western segment to nearly pure strike slipping on its eastern segment. (3) Slip partitioning from oblique thrusting to purely strike slipping from west to east associated with the unique fault geometry of the northeastern Tibetan edge may play an important role on active extension between the Alashan and the Ordos blocks. (4) The present-day surface motion of the northeastern Tibet and surrounding region reflects the active tectonics more likely as rigid-block deformation, especially out of the northeastern Tibetan plateau.
T11C-1277 0800h
Validating Three-dimensional Velocity Models in China and East-Asia for use in Regional Seismic Event Location
We have developed a three-dimensional (3D) a priori geophysical model for the China/East Asia (CEA) region and are using it to construct 3D travel-time tables for seismic event relocation. Validation of this 3D model involves comparison of travel times to other 3D models, both a priori and tomograhic, as well as to observed arrival times from ground truth events. The CEA a priori model is constructed by patching one-dimensional velocity models from previous studies together and adjusting them to conform to known basin and crust thickness values. This type of model aids in the view of geophysical properties over a larger region than typical tomographic models can provide. Using several ray-tracing algorithms, we will produce fine-scale travel-time tables for various stations within the CEA model and compare the travel times to both tomographic models and other a priori models. The 3D travel-time tables will be used to relocate ground truth events for models and determine how these tables affect the regional locations. 3D models do display significant variation in velocities when compared to global one-dimensional models, especially for crust and upper-mantle structure. Our goal is to utilize these 3D variations and apply the 3D travel times on large portions of a seismic catalog to determine their effect. The use of 3D travel-time tables is still being investigated as a viable, operational alternative to one-dimensional models with corrections. As model resolution and accuracy improve, their use can continue to benefit regional event relocation results, but the trade-off with computational load and efficiency must be weighed for the more detailed parameterizations.
T11C-1278 0800h
Plate Tectonics: From Plate Boundary Zone Tectonics To Extensive Intraplate Tectonics
Plates makes up earth's surface, and tectonic activity is generally concentrated on plate boundary zones. In restrict meaning, plate tectonics of the earth is regarded as mixture of plate boundary zone tectonics and extensive intraplate tectonics. For example, the Asian continent never behaves as rigid plate that was deformed extensively when the Indian continent collided with it. I infer that extensive intraplate tectonics reflects rheological weakening of wet mantle. To demonstrate effect of H$_{2}$O component on plate strength, one-dimensional rheological profiles of 100 km depth were constructed by assuming 20km thick upper crust and 20km thick lower crust. Temperature-depth profiles were calculated based on one-dimensional steady-state static heat transfer at given surface heat flows. Power law creep and Byerlee_fs law were used to estimate strength in ductile regime and brittle regime respectively. Creep strength for upper crust, lower crust, dry mantle and wet mantle were calculated using creep parameters of granite, granulite, dry dunite and wet dunite. The minimum value between power law creep strength and Byerlee_fs law strength gives the strength of the lithosphere. Strength profile at surface heat flow of 55mW/m$^{2}$ (continental average is 56.5mW/m$^{2}$) and strain rate of 10$^{-15}$/s (intraplate deformation is about 10$^{-15}$/s - 10$^{-16}$/s in Asia) shows a significant difference in strength for using dry mantle and wet mantle. In case of dry mantle, the uppermost mantle is quite strong. However, if wet peridotite represent the upper mantle, there is very little strength in the uppermost mantle. The cumulative lithospheric strength, i.e. integral strength from surface to 100km depth, and the cumulative mantle strength, i.e. integral strength from 40km to 100km depth were calculated with changing strain rate. For example, to deform continental lithosphere at strain rate of 10$^{-15}$/s, wet mantle has a cumulative strength of about 2x10$^{12}$N/m whereas the cumulative strength of dry mantle is about 15x10$^{12}$N/m. The weakening of wet mantle, therefore, result in the dramatic reduction of cumulative strength of lithosphere from 17x10$^{12}$N/m (dry) to 3x10$^{12}$N/m (wet). Forces of ridge push appear to be in the range of 3.9x10$^{12}$N/m. Ridge push is probably the principal driving force for intraplate deformation after amalgamation of continent. Dry upper mantle carries much of the ridge push force. The magnitude of cumulative lithospheric strength is much higher than that of wet lithosphere and far exceeds the ridge push forces. In contrast, the cumulative strength of wet lithosphere is comparable to the ridge push force, suggesting ridge push forces are enough to cause intraplate deformation of wet continental plate. I infer that slab dehydration along numerous subduction zones during collision of microcontinent and island arcs at late Paleozoic/Mesozoic caused extensive wetting of uppermost mantle beneath Asia, resulting in extensive weakening of plate. Since Proterozoic/Phanerozoic boundary, collision events such as East African orogeny, Variscan orogeny, Alpine orogeny and Himalayan orogeny have involved relatively extensive intraplatel tectonics in Gondwana supercontinent, Pangea supercontinent, Europe continent and Asia continent respectively whereas Archean/Proterozoic continental collisions give rise to plate boundary zone tectonics. It probably reflects initiation of return flow of H$_{2}$O component into mantle at about 750 million years ago (Maruyama, 1999). I conclude that fragile continental plate have been established during late Proterozoic to Phanerozoic assembly of continent, and extensive intraplate tectonics started to operate on earth at Proterozoic/Phanerozoic boundary.
T11C-1279 0800h
How Is Lower Crust Modified As A Neo-Rift Becomes A Paleo-Rift and Part Of The Craton?
The Southern Oklahoma Aulacogen (SOA), at the southern end of Laurentia (present coordinates), if behaving as neo-rifts, such as the Rio Grande Rift, presumably possessed a rift structure in the Cambrian with a continental thickness of about 28km. Seismic data, though sparse, suggest a present thickness of the SOA is about 45km, indistinguishable from adjacent rifted Proterozoic crust. By what process do we add 15km to the original SOA crust: underplating, eclogite-gabbro transformation, or deformation? This question has bearing on how we understand and interpret all paleo-rifts now a part of continental cores. Geology of the southern Midcontinent of North America does not show evidence of significant thermal events in the Phanerozoic. This effectively rules out underplating and phase transformation as a cause of change in M-discontinuity depth. Present SOA outcrops are in the Wichita Mountains of southwestern Oklahoma, part of the easternmost Ancestral Rockies. These outcrops are in the Wichita-Amarillo crustal block uplifted about 7km in the Pennsylvanian. The Anadarko Basin to the north went down about 7km. Large Pennsylvanian thrust faults in the upper brittle crust are documented. Thus it appears that compressive deformation may be able to account for the change in crustal thickness from neo-rift type to paleo-rift and craton type. However, the accommodation made in the lower crust may be more dramatic than deformation in the upper crust because shortening, and thickening of the order of 2X, is probably required. Comparisons with other paleo-rifts in North America, such as the Middle Proterozoic Midcontinent Rift and the NeoProterozoic Reelfoot Rift, show that their crustal thicknesses now also match their previously rifted margins. Can the same sequence, as seems to be the case with the SOA, apply to other paleo-rifts?
T11C-1280 0800h
Large Paleo Landslides Along the Western Part of the Gobi-Altay Fault System in Southwestern Mongolia
A sequence of paleo landslides at the Namalzah Hills, $\sim$70 km south of the town of Altay in southwestern Mongolia (45.8\deg N, 96.5\deg E) is associated with tectonic activity along the western part of the Gobi-Altay Fault system (GAFS). Three mobilized blocks of 0.5, 2.5 and 110 km$^{2}$ suggest multiple events of sliding, and displaced alluvial fans across an adjacent fault trace at the front of the mountain range indicate left-lateral offset. The 110-km$^{2}$ block has been translated $\sim$4.5 km down-slope north from the mountain range, with prominent scarps defining both the eastern and western boundaries of the landslide. Neogene deposits unconformably overlain by Quaternary alluvial sediments up to 200 m thick in places comprise this block, which is structurally characterized by a set of internally drained basins trending east-west, and corresponding terminal lake beds. Well-developed desert pavements characterize its surface. The 0.5- and 2.5-km$^{2}$ blocks, which lie between the 110-km$^{2}$ block and the source area, appear to be younger and thus suggest sliding events that postdate the mobilization of the large block. Elevated alluvial fans found along the mountain front indicate significant antithetic uplift north of the mountain-front fault trace as well as $\sim$2 km of cumulative left-lateral offset. Surface-composition mapping of the largest block suggests 1.0-1.5 km of left-lateral offset between it and the mountain range, while westward translation of the smallest mobilized block indicates $\sim$0.6 km of post-sliding, left-lateral offset. OSL samples were collected from the bottom of a lake bed on the largest block and from the underlying alluvial sediments to provide age constraints for the initiation of these sliding events. The good preservation of carbon recovered from the bottom of the lake bed suggests that the lake is relatively young. Accordingly, slip-rates higher than the 1.2 mm/yr constrained by Ritz et al. (1995) along the eastern part of the GAFS, may be required to accommodate the 1.0-1.5 km of inferred offset between the largest block and the mountain range. While another landslide of similar magnitude has been described by Philip and Ritz (1999) $\sim$400 km to the east along the GAFS, the well-preserved sequence of mobilized blocks and closely related offset alluvial fans of the Namalzah Hills offers a good opportunity to improve our understanding of Quaternary displacement along this part of the GAFS, as well as study the complex relation between tectonic activity and landsliding in such intra-continental environments.
T11C-1281 0800h
Tectonic evolution of the Koshiki Islands at Kyushu in Japan: Extentinal tectonics of the eastern Asian margin at Cenozoic
The Koshiki islands contain one of the best-preserved exposures of upper Cretaceous to Paleogene sedimentary successions in western Kyushu of Japan. These stratified sedimentary sequences also contain well-preserved deformations, used to identify surface conditions and tectonics of the Eastern part of the Asian continental margin. The upper Cretaceous Himenoura Group is approximately 800m thick in this area. It consists of well-rounded conglomerate, cross-bedded sandstone, and black shale and forms coarsening-upward sequence with HCS (hummocky cross stratification) containing many upper cretaceous fossils. This group had been formed in an offshore environment. The Eocene Kamikoshikijima Group, approximately 1700m thick, unconformably overlies the Himenoura Group. It consists of quartz rich cross-bedded sandstone, fine laminated greenish siltstone, and red shale, which are identified as fluvial sediment. Two types of faults are identified; 1) type1: NW-SE trending normal faults composed of thick shear zone with brittle-ductile fabric, and 2) type2: NNE-SSW trending normal faults with sharp fault planes with dip-slip slickenside. The type1 faults have 10m to more than 100m of displacement. Kanoko Fault, which is the biggest type1 fault in this area, has a 5m wide shear zone containing highly foliated shales with dip-slip lineations and asymmetric fabrics. This fault was affected by top to the northeast movement. Based on the comparison between each group stratigraphy, approximately 500m displacements are estimated. We determined the two K-Ar clay mineral ages in the type1 fault gouges. One, which is the Kanoko Fault, dated 16.5-19.1 Ma in age, and another fault, which is deeper situation in Himenoura Group, is 20.6-24.1 Ma in age. The temperature estimation from the fluid inclusions within lineated calcite veins in Kanoko Fault shows that the veins were formed at more than 166?C. Type2 faults contain thin fault gouge with slickenside. These faults clearly truncate the type1 faults. The NNE orientation of the type2 faults is almost parallel to the Koshiki islands and the northeastern Okinawa trough. Three deformations and one intrusive event are identified in this area; 1) tilting of the Himenoura Group (D1), 2) NW-trending normal faults (24-16.5Ma)(D2), 3) intrusion of igneous rocks (14.0+-1.6Ma) and 4) NNE-trending normal faults (D3). During early to middle Miocene in southwest Japan, regional extended stress filed occurred, and D2 to D3 deformations in Koshiki islands might correspond to the changed regional stress filed.