Yves Gaudemer

Active thrusting and folding in the Qilian Shan, and decoupling between upper crust and mantle in northeastern Tibet

Fieldwork south of the city of Gaotai (Gansu province, China) shows that active shortening of surface sediments in the foothills of the Yumu Shan, a large fore-mountain of the Qilian Shan, at the northeastern edge of Tibet, involves both overthrusting and flexural-slip folding. North of this mountain, we found and mapped a prominent north-facing thrust scarp that offsets a Holocene fan sloping gently (3.4°) to the north. Part of this scarp appears to be related to the M ≈ 7.5, 180 A.D. earthquake that may have led to the demise of the Han Dynasty city of Luo Tuo Chen, in the Hexi corridor. A set of 10, 100–150 m long profiles measured across this scarp, 3.2 m high on the average, can be made to fit the diffusion-degraded morphology of a surface break related to the 180 A.D. event using a value of about 3.3 m^2/10^3 yr for the mass diffusivity ϰ of fanglomerates in this part of Gansu province. Smaller mountain-facing scarps on a terrace-capped foothill result from bedding slip concurrent with active folding of underlying, steeply northdipping, Plioquaternary sandstone and conglomerate beds. Holocene uplift rates along the Yumu Shan, which is only one of the Qilian Shan ranges, are estimated to be between 0.4 and 1.9 mm/yr, which implies that much of the mountain formed in the Quaternary. The periclinal structure of the Plioquaternary envelope under which the Paleozoic core of the Yumu Shan plunges towards the west suggests that the whole 3200 m high mountain is a basement ramp anticline. Mountains striking parallel to the Yumu Shan, with similar structure and comparable or greater sizes north and south of the Hexi corridor probably also correspond to recent, crustal ramp anticlines. This implies that the wide, mountainous upper crustal wedge making the northeastern edge of the Tibet-Qinghai plateau is detached from the underlying lower crust and upper mantle.

Northeastward growth of the Tibet plateau deduced from balanced reconstruction of two depositional areas: The Qaidam and Hexi Corridor basins, China

We address the problem of late Cenozoic uplift, erosion, and growth of northeastern Tibet by reconstructing, from isopach maps and drill holes, the volumes and masses of sediments deposited in the Qaidam and Hexi Corridor basins since ∼35 Myr ago. The mass budget is based on simple geometrical assumptions such as regional similarity of the thickness ratios between strata of different ages. In the Qaidam, where our record extends back to the Oligocene, the budget shows a huge rise of the accumulation rates after the beginning of the Pliocene (5.3 Ma). The early Pliocene seems to be the period of maximum deposition with accumulation rates in excess of 1 mm yr−1 (∼ 2.7 kg m−2 yr−1) of compacted rocks throughout the basin. There also seems to be a southeastward shift of the largest depocenters between the upper Pliocene (3.4–1.6 Ma) and the Quaternary. In the Hexi Corridor, sedimentation is confined to small foreland flexural depressions associated with the frontal thrusts of the Qilian Shan and occurs at an average rate one order of magnitude smaller than in the Qaidam basin. The accumulation rate is maximum in the Quaternary. The sedimentation history appears to support a plateau-building mechanism resulting from the combination of two geologicallly common processes: crustal-scale thrusting and sedimentary basin infilling. The time needed to completely fill the Qaidam basin and make its catchment a plateau closely resembling that of the highest part of Tibet (Qangtang) is of the order of 9 Myr. The mechanism now at work north of the Kunlun, which involves rapid infilling of broad, flat areas separated by relatively narrow mountain ranges, has thus probably been important in producing the high, smooth topography that characterizes much of central Tibet.

Holocene left-slip rate determined by cosmogenic surface dating on the Xidatan segment of the Kunlun fault (Qinghai, China)

Cosmogenic dating, using in situ {sup 26}Al and {sup 10}Be in quartz pebbles from alluvial terrace surfaces, constrains the late Holocene slip rate on the Xidatan segment of the Kunlun fault in northeastern Tibet. Two terrace risers offset by 24 {+-} 3 and 33 {+-} 4 m, having respective ages of 1799 {+-} 388 and 2914 {+-} 471 yr, imply a slip rate of 12.1 {+-} 2.6 mm/yr. The full range of ages obtained ({le}22.8 k.y., most of them between 6.7 and 1.4 k.y.) confirm that terrace deposition and incision, hence landform evolution, are modulated by post-glacial climate change. Coupled with minimum offsets of 9--12 m, this slip rate implies that great earthquakes (M {approximately}8) with a recurrence time of 800--1000 yr, rupture the Kunlun fault near 94 E.

Primary surface ruptures of the great Himalayan earthquakes in 1934 and 1255

It is unclear where plate boundary thrusts generate giant rather than great earthquakes. Along the Himalayas, the source sizes and recurrence times of large seismic events are particularly uncertain, since no surface signatures were found for those that shook the range in the twentieth century. Here we challenge the consensus that these events remained blind and did not rupture the surface. We use geomorphological mapping of fluvial deposits, palaeo-seismological logging of river-cut cliffs and trench walls, and modelling of calibrated 14C ages, to show that the Mw 8.2 Bihar–Nepal earthquake on 15 January 1934 did break the surface: traces of the rupture are clear along at least 150 km of the Main Frontal Thrust fault in Nepal, between 85 50 and 87 20 E. Furthermore, we date collapse wedges in the Sir Valley and find that the 7 June AD 1255 earthquake, an event that devastated Kathmandu and mortally wounded the Nepalese King Abhaya Malla, also ruptured the surface along this stretch of the mega-thrust. Thus, in the past 1,000 years, two great earthquakes, 679 years apart, rather than one giant eleventh-century AD event, contributed to the frontal uplift of young river terraces in eastern Nepal. The rare surface expression of these earthquakes implies that surface ruptures of other reputedly blind great Himalayan events might exist.

Uniform slip‐rate along the Kunlun Fault: Implications for seismic behaviour and large‐scale tectonics

A long-term slip-rate is derived from concordant 10 Be, 26 Al and 14 C dating of cumulativeoffse ts along much of the length of the Kunlun Fault. Values at 6 sites indi- cateuniform slip (11 .5 ± 2.0 mm/yr) since ∼40 kyr BP. This relatively high slip rate corresponds to a first-order discon- tinuity in the Asian crustal velocity field. M∼ 8a nd M∼7.5 earthquakes on 2 segments of the fault recur with charac- teristic slip (∼10 ± 2ma nd 4.4 ± 0. 4m ) ev e ry∼850 and ∼420 yrs, respectively.

Slip accumulation and lateral propagation of active normal faults in Afar

We investigate fault growth in Afar, where normal fault systems are known to be currently growing fast and most are propagating to the northwest. Using digital elevation models, we have examined the cumulative slip distribution along 255 faults with lengths ranging from 0.3 to 60 km. Faults exhibiting the elliptical or “bell-shaped” slip profiles predicted by simple linear elastic fracture mechanics or elastic-plastic theories are rare. Most slip profiles are roughly linear for more than half of their length, with overall slopes always <0.035. For the dominant population of NW striking faults and fault systems longer than 2 km, the slip profiles are asymmetric, with slip being maximum near the eastern ends of the profiles where it drops abruptly to zero, whereas slip decreases roughly linearly and tapers in the direction of overall Aden rift propagation. At a more detailed level, most faults appear to be composed of distinct, shorter subfaults or segments, whose slip profiles, while different from one to the next, combine to produce the roughly linear overall slip decrease along the entire fault. On a larger scale, faults cluster into kinematically coupled systems, along which the slip on any scale individual fault or fault system complements that of its neighbors, so that the total slip of the whole system is roughly linearly related to its length, with an average slope again <0.035. We discuss the origin of these quasilinear, asymmetric profiles in terms of “initiation points” where slip starts, and “barriers” where fault propagation is arrested. In the absence of a barrier, slip apparently extends with a roughly linear profile, tapered in the direction of fault propagation.

Postglacial left slip rate and past occurrence of M≥8 earthquakes on the Western Haiyuan Fault, Gansu, China

High-resolution (HR) air photographs and a 1-m horizontal and 2-m vertical resolution digital elevation model derived from them by stereophotogrammetry provide new constraints on the behavior of the western stretch of the active Haiyuan fault, in Gansu province, China. The photographs cover three swaths along the fault, each about 2-km-long and at least 500-m-wide, near the village of Songshan, at 103.5oE. This high-resolution data set is used to map and measure cumulative horizontal offsets of alluvial terraces and risers that range between 115 and la 135 m, and 70 and 90 m, at two sites. Dating these terraces with 14C yields minimum and maximum ages of 8400 and 7600, and 14,200 years B.P., respectively. This leads to a postglacial slip rate of 12±4 mm/yr, with a most likely minimum value of 11.6±1.1 mm/yr. The smallest stream offsets observed on the HR photographs range between 8 and 16 m and are interpreted as coseismic displacements of the last few earthquakes with M ≥ 8 that ruptured the 220-km-long Tianzhu gap of the fault, west of the Yellow River. Earthquakes of that size within this gap, which has been quiescent for at least 800 years, would recur at intervals of 1050±450 years.

Thermal control on post-orogenic extension in collision belts

Abstract Intracontinental extension develops in two different ways. In the Rhine Graben, the Baikal or the East African Rifts, it is localized in a long and narrow zone, along a few normal faults which often delineate a single graben. In contrast, in the Basin and Range Province or the Tibetan Plateau, it is distributed over a much broader area, up to more than 1000 km across. This difference arises from differences in rheology and applied forces. Widespread extension probably reflects the thinning of a previously thickened crust. Geological observations indicate that only wide orogenic belts undergo late or post-orogenic thinning. A related fact is that anatectic granites seem to be most abundant in wide orogenic belts. The general correlation between belt width and occurrence of both extension and anatectic granites suggest that it reflects in part the characteristics of horizontal heat transfer between thickened crust and surroundings. In wide orogenic plateaux, the large amount of heat produced in the crust by radioactive decay and lack of erosion leads to high temperatures. In narrow mountain belts, horizontal heat transfer and high erosion rates concur to prevent temperatures from rising to values at which the crust may yield in extension. We have tested this hypothesis using transient two-dimensional numerical models of orogenic belts, focussing on the effects of the shortening rate, the final width of the belt and the rate of crustal radiogenic heat production. In the range of geological values, the shortening rate has little influence on the results. Crustal temperatures increase as a function of belt width. We find that the sensitivity of crustal temperatures to belt width is most pronounced in the 100–300 km width range and for large values of crustal heat generation. The pressure-temperature paths followed by geological formations depend on the belt width, which is consistent with available data. The integrated shear strength of the lithosphere decreases with time as the belt is being built and then eroded. This evolution is sensitive to belt width. If extension occurs when the integrated shear strength drops below a critical value, our results yield a framework for comparing different orogenic belts. We predict that extension starts late in the history of wide belts and never occurs in narrow belts. A final implication of this study is that the average rate of crustal heat production is probably close to 1.1 × 10 −6 W/m 3 , which is in the upper range of possible values. We present a simple and convenient diagram summarizing the thermal characteristics of orogenic belts as a function of width and shortening rate.

Ductile and brittle deformations in the northern snake range, nevada

Abstract The northern Snake Range, east-central Nevada, is one of the metamorphic core complexes of the Sevier hinterland. Within the range a major decollement separates an ‘upper plate’ composed of brittlely deformed Paleozoic sedimentary rocks (mostly carbonates), from a ‘lower plate’ composed of metamorphic Upper Precambrian-Lower Cambrian rocks, intruded by gneissic granites. A study of the geometry and kinematics of structures and fabrics at outcrop scale and in thin sections indicates that: the northern Snake Range decollement has been a zone of intense non-coaxial E-vergent shear and transport in a ∼- N115°E direction. Outstanding asymmetric boudinage within the marble sheet capping the lower plate testifies for late ductile shear strains (γ) of at least 20. The interface between brittlely and ductilely deformed rocks seems too sharp to represent a regional rheological transition, but might result from two distinct phases of deformation. Ductile deformation in and below the decollement could have occurred before brittle deformation in the upper plate. Brittle faulting in the upper plate related to Basin and Range extension reactivated the upper surface of the ductile shear zone. High topographic relief on the normal faults bounding the range triggered the slide of olistolites from the upper plate into the adjacent Oligo-Miocene basins.

Geometry and rate of faulting in the North Baikal Rift, Siberia

We present a detailed morphotectonic analysis of late Quaternary faulting in the North Baikal Rift (NBR), a region characterized by ranges and basins distributed over more than 800 km along strike in eastern Siberia. Remote sensing techniques (SPOT, METEOR scenes, and aerial photographs) are used to map the active fault network which displays a general en echelon distribution from the northern Lake Baikal to the easternmost basin, with ∼30-km-spaced overstepping segments of 10–80 km in length. Most faults have a dominant dip-slip component over their Cenozoic history. The inherited crustal fabric strongly influences the overall geometry of the rifted basins. We use 54 14C ages of postglacial terraces near the foot scarps of the Muya basin to date offsets measured inside alluvial fans. The last main postglacial event in this area appears to be the early Holocene optimum dated at ∼10 ± 2 ka, following the onset of deglaciation at ∼13 ka. Using these time constraints, a detailed leveling across two terraces offset by the Taksimo fault (West Muya basin) shows consistent minimum vertical slip rates of 1.6±0.6 mm yr−1. Using 30 other active scarps analyzed in the field, we find a lower bound for horizontal velocity of 3.2±0.5 mm yr−1 across the NBR, a rate close to the one found in the southern rift from Global Positioning System measurements. We then compare directions of slip vectors from Holocene field data and slip directions from earthquake fault plane solutions: although local discrepancies appear, the mean directions of lesser horizontal stress (σ3) inverted from theses values are ∼N130°E and ∼N155°E, respectively, which are comparable within uncertainties and favor a rifting obliquity of ∼30°–40°. Extrapolating our Holocene rates, we estimate basin ages younger than those generally believed (less than 7 Ma) and propose a spatial and temporal evolution of rifted basins consistent with experimental models of oblique rifting. Total amounts of extension and vertical throw (∼7 and ∼12 km, respectively) across major faults appear rather constant from the central to the northern rift. These results favor a progressive development of asymmetric grabens in a rift zone that widens with times and they indicate a strong rheological control on deformation which seems enhanced by other contributions than the far-field effects of the Indo-Eurasian collision.