T34C-01 INVITED
Erosion-Induced Reactivation of the Main Central Thrust zone: Model and Implications for Channel Flow in the Himalayan-Tibetan System
Recent thrust-sense deformation in the vicinity of the Main Central Thrust (MCT) zone in the Himalaya of central Nepal can be attributed to tectonics, erosion, or a combination of both. In the same area, contrasting cooling-age patterns in medium- and low-temperature thermochronometers have been interpreted to imply a recent (2.0-0.9 Ma) significant increase in erosional exhumation rates, likely attributable to recent climate changes. No comparable evidence exists for changes in local plate convergence rates. We use numerical models with constant convergence velocity but contrasting erosion rates to show that increased erosion and recent thrusting may be directly connected. In the models, increasing erosivity by a factor of 3 over 3 Ma fundamentally changes the style of deformation, reactivating the dormant model MCT system in the region corresponding to observed thrust faults. The high-erosion model also reproduces the observed cooling-age patterns, whereas the equivalent low-erosion model does not reproduce either observation. Other model predictions, and their implications for the effects of increased erosion on the southern flank of the Himalaya, include: 1) no associated reactivation of normal faulting on the South Tibetan Detachment (STD) system; 2) enhanced upper-crustal extension in the vicinity of the north Himalayan gneiss domes (NHGD); 3) re- invigorated mid-crustal channel flow beneath the NHGD; 4) possible destabilization and wholesale southward flow of the upper crust between the MCT and NHGD, with the potential for catastrophic earthquakes. The first three predictions are testable and address the persistent question of the existence and current location of the low-viscosity channel. In particular, the model predicts that it has been stagnant beneath the Tibetan plateau under a relatively low-erosion regime, but has been, or could be, reactivated by more aggressive erosion driven by climate change. The potential for destabilization of the south flank of the orogen has important geological and societal implications.
T34C-02 INVITED
Is rapid exhumation of the High Himalaya driven by ramp overthrusting, the formation of mid- crustal duplex, or out-of-sequence thrusting associated with channel flow?
A number of studies of river incision and exhumation across the Nepal Himalaya indicate that uplift rates must increase abruptly from the Lesser Himalaya to the High Himalaya. Locally high uplift rates could relate to: 1- thrusting over a mid-crustal ramp; 2- mid-crustal duplex growth; or 3- out-of-sequence thrusting along the front of the High Himalaya. Ramp-overthrusting can be a viable mechanism only over a short period of time since, considered alone, this mechanism fails to account for the growth of the orogenic wedge. The same problem arises with out-of-sequence thrusting, if considered alone. Thus an additional mechanism is needed to account for the transfer of material from the Indian crust to the Himalayan wedge. The structure of the Himalayan wedge as well as the pattern of active deformation across the range suggests that underplating, through the development of a duplex system at mid-crustal depth, has been the dominant mechanism of accretion over the last ~15Myr. In an attempt at determining plausible kinematic models over this period, we have applied a formal inverse approach based on the Neighbourhood Algorithm. The forward models consider the possibility of either thrusting localized along a single major thrust fault (the MHT) with non-uniform underplating due to duplexing, or out-of sequence thrusting in addition to thrusting along the MHT and with uniform underplating rate. The models are computed using a thermokinematic FEM model (PECUBE), and tested against thermochronological, thermometric and thermobarometric data from central Nepal compiled from the literature and complemented with new (U-Th)/He data. The formal inversion approach allows definition of best fitting values of model parameters and their uncertainties. In addition to the geometric parameters, model variables include overthrusting rates; radiogenic heat production in the High Himalayan Crystalline (HHC) sequence; the timing of enhanced rock uplift /exhumation rates corresponding to the formation of the duplex or out-of-sequence thrust re-activation. A model with out-of-sequence thrusting can provide a satisfactory fit to the data (with a minimum reduced χ2 of 1.41) but requires a large overthrusting rate of 11 +- 1mm/yr and implies a total convergence rate >30 mm/yr. The duplex model, with a minimum reduced χ2 of 1.01, is more consistent with observation. According to this model, the 20 mm/yr convergence results from an overthrusting rate of 5 +- 1 mm/yr and an underthrusting rate of 15 +- 1 mm/yr. Modern uplift rates are estimated to increase from about 0.8+-0.2 mm/yr in the Lesser Himalaya to 3.5 +- 0.5 mm/yr at the front of the high range, 95 +- 5 km from the MFT. The effective friction coefficient is estimated to be 0.09 +- 0.01 and the radiogenic heat production of HHC units is estimated 2.2 +- 0.1 μW/m3. The mid-crustal duplex initiated at 10 +-2 Ma, leading to an increase of uplift rate at front of the High Himalaya from ~0.8 to 3.5 mm/yr. Thus the analyzed dataset appears more consistent with a duplex model relative to out-of-sequence-thrusting. Given the excellent fit of the duplex model to the geologic, petrologic and geophysical constraints, there appears little basis to support a channel flow-type model in a Himalayan context over the last 15Myr.
T34C-03 INVITED
Unchannelized collapse of thickened continental crust and metamorphic core complex formation
Features common to a subset of metamorphic core complex footwall rocks globally include ductile fabrics, migmatitic rocks, mid-crustal maximum pressures, spatial and temporal association with orogenic crustal thickening, spatial association with lateral variations in crustal strength, and association with brittle normal faults. Continuous deformation of fluids under boundary conditions that allow finite vertical and horizontal velocities while maintaining continuity of vertical stresses through the lithosphere provide one candidate mechanism for formation of these complexes. Such systems simultaneously produce mid-crustal rock uplift by gravitational collapse and large extensional stresses at the base of an upper-crustal elastic lid. We explore the application of unchannelized flow models to case studies in the northern Rockies of the U.S. and the eastern Pamir.
T34C-04 INVITED
Channel Flow in the Deep Crust and its Topographic Expression
Crustal channel flow occurs when lateral flow in the mid to deep crust occurs more rapidly or in a different direction than surface motions, measured with respect to the underlying mantle. Because channel flow accomplishes lateral transfer of crustal material with little development of surface geologic structures, documenting and constraining channel flow remains a challenge. Channel flow may transfer lower crustal material over length scales of hundreds up to perhaps a thousand kilometers. It can occur along strike, such as in the central Andes, across strike, such as may have occurred across the early Paleozoic passive margin of western North America; it always occurs down the regional topographic gradient. Channel flow requires more than the existence of a weak lower crust. Either the lower crust must be weak enough (low enough viscosity) that flow does not cause stresses in the upper crust to exceed the brittle yield criteria, or, if not, it occurs only if the ratio of upper crustal to lower crustal viscosity exceeds a value that scales linearly with the flow-length squared and with the channel thickness. Channel flow on length scales of 100 km is easily developed, while channel flow on length scales of 1000 km should be relatively rare. Because a deep crustal channel is effectively a crustal asthenosphere, the conditions necessary for channel flow ensure that the regional topographic gradient above the channel is very low. It thus provides a natural mechanism for establishing areas of low relief at high elevation that is independent of the density structure within the underlying mantle. Probably most low relief, high elelvation regions on Earth are underlain by areas of channel flow, including Tibet, Altiplano and much of the high central Andes, the Iranian Plateau, and the high areas of eastern Turkey.
T34C-05
Quantifying landscape differences across the Tibet plateau: Implications for topographic relief evolution
Because of its extreme high-elevation and low-relief interior with a particularly large areal extent, the Tibet plateau is a unique topographic feature on Earth. How and when it attained its current topography has been the motivation of many studies. In one class of previous studies, the landscape morphology of the plateau has been taken to be a straight-forward reflection of deep seated lower crust and mantle processes. The importance of surficial erosion processes in shaping the landscape, by contrast, has been downplayed. To assess the potential role of such processes, we quantify here the bulk landscape characteristics in representative regions of northern, central and southeastern Tibet. We find that the northern and central parts of the plateau are characterized by positive topography. Internal drainage appears to play a major role in smoothing out the tectonically generated structural relief. The change from the internally drained interior, to the externally drained part of eastern Tibet is accompanied by a transition from low to high relief, and from positive to negative topography. We propose that the evolution of river systems and drainage efficiency coupled with tectonic uplift provides a robust mechanism in explaining the systematic regional differences in Tibetan landscape. This also provides a unifying mechanism for the formation of the low-relief interior, and for the origin of the high-elevation low-relief relict surface in SE Tibet. Our analysis questions the fashionable contention that a continuous pre-uplift low-relief surface first formed at low elevation, extending all the way to the South China Sea shore, before being warped upwards in the Late Miocene-Pliocene by lower crustal channel flow.
T34C-06
U(-Th)-Pb age Constraints on the Timing and Duration of Channel Flow in the Mt. Everest Region, Eastern Himalaya
U(-Th)-Pb dating of accessory phases from metamorphic and igneous rocks at two outcrops along a north-south transect in the Mt. Everest region of southern Tibet provides new constraints on the timing and duration of thermal events associated with ductile extrusion of the Greater Himalayan Series (GHS). At the southern-most outcrop in the Kangshung valley, Th-Pb ages from monazite indicate that prograde metamorphism associated with crustal thickening following the India - Asia collision, peaked at least as early as ~39 Ma, ~7 Ma earlier than previously recognized in the GHS of southern Tibet. A subsequent sillimanite grade metamorphic event at ~28 Ma was followed by two phases of leucogranite emplacement at ~21 Ma and ~17 Ma. At Thongmön, 40 km in the down-tectonic transport direction relative to Kangshung valley, prograde metamorphism was occurring at ~25 Ma and lasted until ~16 Ma, reaching ~740°C and 5 kbar at ~22 Ma. Immediately following metamorphism, two phases of leucogranite were emplaced at ~15 Ma and ~12 Ma, with an intervening phase of ductile deformation. These two outcrops record a northward propagation of metamorphic, magmatic and structural events that are ~5-7 Ma younger in the down-tectonic transport direction. This diachroneity in thermal history provides quantitative geological support for previously proposed models of south-directed mid-crustal channel flow, but suggests that the time-scales over which this process operated require significant revision.
T34C-07
Empirical Constraints on Extrusion Mechanisms Derived From Pressure-Temperature-Time Histories From the Himalayan Metamorphic Core (Sutlej Valley, NW India)
The exhumed Himalayan core in the Sutlej Valley comprises the Greater Himalayan Sequence (GHS), bounded by the Main Central Thrust and the South Tibetan Detachment, and tectonically distinct metamorphosed units above (the Haimanta Group) and below (the Jutogh Group). While pressure-temperature-time (P-T-t) data from the GHS are broadly compatible with predictions of the channel flow model presented by Jamieson et al. (2004), corresponding data for the units bounding the putative channel are not: The underlying Jutogh Group experienced a tightly closed P-T path featuring upper-amphibolite prograde metamorphism at c. 11 Ma, followed by rapid cooling and exhumation. These data are consistent with a) prograde metamorphism during overthrusting (along the Main Central Thrust) and b) subsequent exhumation via accretion to the extruding GHS channel above. However, muscovite cooling ages from the GHS pre-date those from the Jutogh Group by at least 10 Ma, clearly indicating decoupled exhumation of the two crystalline units. Alongside evidence that motion on the Main Central Thrust had ceased by c. 16 Ma, concomitant extrusion of the Jutogh Group as part of a single widening GHS `channel' seems impossible. The Haimanta Group, considered uppermost in the GHS sequence and/or basal to the Tethyan Sedimentary Sequence, also reached upper-amphibolite grade during the Himalayan orogeny. As for the Jutogh Group, exhumation and cooling rapidly followed peak metamorphism. Coupled U-Pb monazite data and detailed textural and pseudosection analyses constrain prograde metamorphism at c. 35 to 30 Ma, the timing of which is compatible with channel flow model predictions. Importantly, however, P-T paths do not match numerical simulations, which imply shallower burial, lower peak temperatures and a distinct phase of isobaric heating. We conclude that the Sutlej Valley presents a tectonically complex metamorphic core for which a single, widening channel flow model does not accurately predict the P-T-t conditions observed in units above and below the proposed channel. Modifications to the model (e.g. migrating focused denudation) may provide more realistic predictions. It also seems likely that tectonic evolution of the Himalaya encompassed aspects of models incorporating both foreland thrust propagation and channel flow, at different stages during its history. Jamieson, R. A., Beaumont, C., Medvedev, S. and Nguyen, M. H. 2004. Crustal Channel Flows: 2. Numerical Models with Implications for Metamorphism in the Himalayan-Tibetan Orogen. Journal of Geophysical Research- Solid Earth, 109, art. no.-B06407.
T34C-08
Did the Karakoram fault interrupt mid-crustal channel flow in the western Himalaya?
There is a marked change in the volume and age of granitoids from west to east across the Himalaya; that change occurs at the southeastern termination of the Karakoram fault where it merges with the Indus-Yarlung suture zone, near the Gurla Mandhata gneiss dome. These granitoids are derived from a ductile mid-crustal channel formed when anatectic melts from the mid-crust beneath the Tibetan plateau were driven south by erosion at the Himalayan topographic front. The "channel flow" model predicts upwellings of these granites within the Tethyan Himalaya Sequence as part of a chain of gneiss domes and exposure of the channel at its southern termination in the Greater Himalaya Sequence (GHS) as evidenced by widespread migmatites and leucogranite bodies at the top of the GHS in the footwall of the South Tibetan Detachment. Leo Pargil is the westernmost gneiss dome in a chain of domes formed within the Tethyan Himalaya Sequence that extends eastward 1600 km through the North Himalayan gneiss domes (that includes the better known Kangmar dome). New U-Pb SHRIMP dating of zircons show leucogranite bodies from the Leo Pargil gneiss dome are Early Miocene (22-20 Ma) corresponding to some of the older granites dated from across the Himalaya. Transects through the GHS along the Beas and Sutlej River valleys in the western Himalaya near Leo Pargil reveal rare to minimal amounts of migmatites and leucogranite, whereas there are abundant migmatites and large leucogranite bodies in a transect along the Friendship Highway from Lhasa to Kathmandu in the area of the North Himalayan gneiss domes. The timing of the initiation of slip on the Karakoram fault is dated from syn- kinematic rocks in the shear zone at Tangste/Pangong Tso at c. 25-21 Ma. If the Karakoram fault acted as a barrier the flow of granitoid melts in a mid-crustal channel, one would expect more abundant and younger granites east of the Karakoram fault as the channel continued to flow south after c. 20 Ma; the youngest granites reported from the eastern Himalaya are 7 Ma from the Renbu dome and there are also abundant leucogranite bodies from 20-12 Ma. This model also predicts that granitoids would leak up the Karakoram fault and this is demonstrated by 20-16 Ma leucogranites along the Karakoram fault in the Tangste/Pangong Tso area. If the Karakoram fault is indeed a barrier to ductile flow and melt migration, then future tectonic models for the Himalaya should recognize the differences between the western and eastern segments of the orogen, rather than emphasizing along-strike uniformity.
T34C-09
Construction of the Eastern Himalaya by Thick-skinned Thrust Stacking of the Indian Basement: No Lower Crustal Flow from Tibet is Needed
A geologic investigation integrating filed mapping, geochronology and geochemistry across the Eastern Himalaya at the longitude of 91-93E has led to the following findings and interpretations. (1) The Triassic flysch complex in the North Indian Sequence (also known as the Tethyan Himalayan Sequence) was sourced from the Lhasa terrane (not India!). Thus, geochemical signatures of Cenozoic intrusive rocks alone in the northern Himalaya are insufficient to test whether the Tibetan lower crust has extruded ductilely below the Himalayan orogen, because India and Lhasa shared the same basement geology prior to the opening of the Neo-Tethys in the Triassic and early Jurassic. (2) The Main Central Thrust warps over a large thrust duplex. The total amount of shortening across the fault and its footwall thrusts exceeds 500 km. This estimate may include contribution from an early Paleozoic contractional event that is at the present difficult to be distinguished from Cenozoic strain. Work is in progress in the Shillong Plateau to resolve this issue. (3) Our U-Pb zircon dating together with the existing geochronologic results indicates that the ages of Eastern Himalayan orthogneisses cluster at 500 Ma, 870 Ma, 1100 Ma, and 1770 Ma, respectively. These ages correlate well with those found in the Shillong Plateau of the Indian basement, immediately south of the Eastern Himalaya. The lithologic correlation between the Himalayan and Indian basement units implies that the construction of the Eastern Himalaya was accomplished by thick- skinned, basement-involved thrusting. Motion on the deeply rooted thrusts has stacked and thus thickened the Indian basement to form the Eastern Himalayan orogen.
T34C-10
Slip-rate gradients along the eastern Kunlun fault: Implications for crustal strength in eastern Tibet
The degree to which convergence between India and Eurasia is accommodated by slip on intra-continental strike-slip faults is central to the question of the strength of Tibetan crust/lithosphere. Although rapid, spatially uniform slip along these structures is often cited as evidence for a strong crust, recent recognition of displacement-rate gradients near the tips of these structures afford an opportunity to evaluate the relationship between fault slip and distributed deformation of the surrounding plateau. Here we present new estimates of slip rate from displaced geomorphic markers at 6 sites along the easternmost ~ 150 km of the Kunlun fault. We reconstruct displacement using high-resolution surveys of displaced landforms and combine these with a regional chronology derived from radiocarbon and cosmogenic isotopes. Collectively, these rate determinations reveal a systematic eastward decrease in slip-rates toward the fault tip. Offset, early Holocene glacial moraines at our westernmost site indicate rates of ~ 7 mm/yr near 99.6° E. These rates appear to decrease eastward to ~ 6 mm/yr at 100.4° E, ~ 5 mm/yr near 101° E, 3-4 mm/yr near 101.5° E, and finally to ~ 2 mm/yr near 102° E. Gradients in slip-rate are broadly matched by geodetic velocities, which show a similar decrease in the far-field shear across the Kunlun fault. If the fault tip has remained relatively stationary during the Holocene, then displacement gradients should reflect the constitutive properties of the crust (e.g., Barr and Houseman, 1996). Preliminary analysis indicates that while the long length- scale of the slip-rate gradient implies a relatively strong crust, the short length-scale over which distributed shear decays away from the fault tip implies a weak crust. Incorporation of our slip-rate data and the far-field surface motion rates into a 2D visco-elastic model should elucidate whether this gradient uniquely requires a fault tip in a strong crust, a weak crust, or neither.