当前位置：
X-MOL 学术
›
Lithosphere
›
论文详情
Our official English website, www.x-mol.net, welcomes your
feedback! (Note: you will need to create a separate account there.)
Lithospheric Deformation and Active Tectonics of the NW Himalayas, Hindukush, and Tibet
Lithosphere ( IF 1.8 ) Pub Date : 2021-01-08 , DOI: 10.2113/2021/7866954 Ishtiaq A. K. Jadoon 1, 2, 3 , Lin Ding 1, 2, 4 , Saif-ur-Rehman K. Jadoon 1, 2 , Zahid I. Bhatti 5 , Syed T. H. Shah 3 , Muhammad Qasim 1, 3
Lithosphere ( IF 1.8 ) Pub Date : 2021-01-08 , DOI: 10.2113/2021/7866954 Ishtiaq A. K. Jadoon 1, 2, 3 , Lin Ding 1, 2, 4 , Saif-ur-Rehman K. Jadoon 1, 2 , Zahid I. Bhatti 5 , Syed T. H. Shah 3 , Muhammad Qasim 1, 3
Affiliation
The Himalayan Mountain System (HMS) and the Tibetan Plateau (TP) represent an active mountain belt, with continent-continent collision. Geological and geophysical (seismological modeling, seismic reflection, and gravity) data is reviewed herein for an overview of the lithospheric deformation and active tectonics of this orogen. Shallow crustal deformation with dominance of thrusting along the margins of the TP is interpreted with normal faulting in the center and strike-slip deformation with the lateral translation of blocks, over a wedge of ductile deformation. The seismicity is the linear concentration over the margins of the orogen to ~20 km depth with exception of the Hindukush and Pamir having seismicity to 300 km depth with an interpretation of sinking Indian and Asian lithospheres. The lithospheric structure is represented by mechanically weak surfaces representing décollement to 15 km depth over the basement, low-velocity zone (LVZ) at ~20 km, the Moho at ~40-82 km, and the lithosphere-asthenosphere boundary (LAB) at 130-200 km depth. The décollement, termed as the Himalayan Mountain Thrust (HMT), is inferred to be rooted at the base of the Moho in central Tibet. Along this fault, brittle crustal deformation is interpreted to ~15-20 km depth, with brittle-ductile deformation along LVZ and ductile slip with crustal duplexing along the lower crust. The mantle lithosphere of the Indian plate is inferred as duplicated with the wedging of the Asian mantle lithosphere. The active tectonics of the TP is proposed to follow the mechanics of thrusting, similar to the foreland deformation of the mountain belts and accretionary prisms.The HMS and the TP represent the world’s highest and biggest active mountain belt, with an advanced stage of the continent-continent collision between Indian and Eurasian plates (Figure 1). The indentation of the Indian into the Eurasian plate since about 55 Ma [1] has led to uplift of the TP to a height of ~5 km with north-south crustal stacking, east-west extension, and lateral extrusion of rigid segmented blocks along strike-slip faults [2–9].Surface geological (such as magmatism, metamorphism, and deformation) [10–12] and subsurface geophysical studies based on gravity modeling [5, 13, 14], seismicity (Figure 2; [2, 5, 6, 15–17]), seismic reflection interpretations [18, 19], and seismic tomography [4, 20–22] have improved our understanding about the structure, deformation, and evolution of the HMS and the TP. Through these studies, foreland and hinterland deformation is better understood, with thin- and thick-skinned deformation over the crystalline basement, along a narrow zone of shallow seismicity in the Himalayan arc [15, 17]. Full-thickness continent-continent (Figures 3 and 4) and ocean-continent transitional crust is modeled with gravity anomalies along the main Himalayas and the western margin of the Indian plate to represent an advance and/or an earlier stage of collision, respectively [13, 14], with the sinking of the Indian/Eurasian lithosphere to a depth of ~300 km in the Hindukush and Pamir [2, 6, 20, 23]. The extent of the Indian lithosphere and its underplating below the Eurasian plate or vice versa is being analyzed and debated [21, 24, 25]. Tomographic studies have added to our understanding about the lithospheric structure (Figure 4), with the interpretation of LVZ at ~20-25 km, the Moho at ~82 km, and LAB at ~200 depth in the TP [4, 21, 22]. In this study, the geological and geophysical data is integrated for review and a better understanding of (1) active tectonics, (2) lithospheric structure and deformation, and (3) evolution of the TP to provide a framework for overall understanding of the orogen and future studies.The geological setting of the NW Himalayas, Karakoram, and Hindukush is unique with junctions of interacting plates and microplates. Tectonic evolution of the TP is associated with the amalgamation of multiple tectonic terrains between Indian and Eurasian plates since the Permian [10]. From north to south, these are Songban-Ganzi, Qiangtang, and Lhasa sandwiched between the Eurasian and the Indian plates (Figure 1). The Songban-Ganzi terrain was accreted to the Eurasian plate along the Kunlun-Qinlin suture in the late Permian. The Qiangtang terrain was accreted to the Songban-Ganzi along the Jinsha suture in the late Triassic-early Jurassic. The Lhasa terrain was accreted to the Qiangtang along the Banggong suture in the late Jurassic, with the final accretion of the Indian plate with the Eurasian along the Zangbo suture in the late Paleocene-early Eocene. The timing of collision along the IZS is variably reported from the Cretaceous (65 Ma) [26, 27] to the Oligocene (34 Ma) [28].The Pamir and the Karakoram represent accreted terrains to the Eurasian plate in north Pakistan separated from the TP by right-lateral, NW-SE oriented Karakoram fault (KF). In the NW Himalayas, the Zangbo suture bifurcates into a northern suture, termed as the main Karakoram thrust (MKT), and a southern suture, termed as the main mantle thrust (MMT), bounding the Kohistan-Ladakh island arcs towards north and south, respectively ([12]; Figure 2). The timing of collision along northern (MKT) and southern (MMT) sutures is recorded as ~90 and ~55 Ma, respectively [1, 11, 29].The HMT and the TP have been extensively studied to understand the ongoing process of continental collision. Interpretation of the industry seismic reflection profiles suggests thin-skinned deformation with the presence of a décollement at 3-10 km depth over the crystalline basement in the Himalayan foreland in north Pakistan [18, 19]. Bouguer gravity modeling suggests full-thickness continental crust of ~38 km from the Himalayan foredeep to the suture zone [13], representing continent-continent collision in the main Himalayas, with a duplicated crust to 82 km thickness in the TP ([21]; Figure 4). On the contrary, the ocean-continent transitional crust, of 15 to 27 km from the hinterland to foreland, respectively, is modeled in the Sulaiman fold belt, along the western margin of the Indian plate [14]. The seismicity in the Hindukush extends to ~300 km depth (Figure 5), along the northwest-dipping Indian lithosphere (Figure 6), contrary to the south-dipping Eurasian lithosphere along north-south profiles across the Pamir with its southern limits along the Rushan-Pshart zone (RPZ) (Figures 6 and 7). Seismic tomographic studies show the lithosphere to have a thickness of ~200 below the TP due to north-south shortening between the colliding plates [21]. GPS slip measurements [9, 30] suggest thrusting along the Himalaya and Altyn-Tagh as the main mechanism of deformation, with north-south oriented normal faulting in the center of the plateau due to east-west extension and strike-slip faulting due to lateral extrusion of rigid blocks [30]. Thin- and thick-skinned deformation is interpreted in the foreland and hinterland of the HMS with shallow crustal deformation [15, 17, 31].Figure 7 shows the south-dipping Asian plate under the Indian plate along the Pamir [20]. Two major tectonic models are proposed in previous studies based on seismicity in the Hindukush and Pamir. The first model suggests two converging seismic zones with the northward subducting Indian plate under the Hindukush and the southward subducting Asian plate under the Pamir [3, 23, 32]. The second model suggests that the intermediate-depth seismicity under the Hindukush and Pamir is caused by a single subducting Indian slab that overturns under the Pamir [33, 34]. A multiple face-to-face subduction model for the evolution of Tibet is widely accepted [35]. The present study is focused on the review of geological and geophysical data for an overall understanding of shallow and deep crustal deformation along the NW Himalayas, Karakoram, and Tibet.General observations and interpretations related to active tectonics and the lithospheric structure of the HMS and the TB are as follows. (1)The HMS is a result of a collision between the Eurasian and the Indian plates, since about 55 Ma [1]. Ongoing collision has led to the duplication of crust under the TP with the Moho at ~70-82 km depth [13, 21](2)Active brittle deformation is mostly confined to ~20 km depth along a narrow zone of deformation of steep topography along the frontal part of the mountain system and along sutures and faults in the internal part of the TP [5, 7, 15, 17, 31, 36]. Intermediate-depth (up to 300 km) seismicity is recorded along the Hindukush and Pamir with an interpretation of the sinking lithosphere of the Indian and the Eurasian plates [6, 20](3)The foreland deformation is interpreted to have a thin-skinned imbricate style of deformation in the Salt Range-Potwar Plateau (SRPP) and duplex style of deformation in the Sulaiman fold belt (SFB), constrained with industry seismic reflection profiles [18, 19, 37, 38] and seismicity [7, 39] (Figures 8–10). The hinterland deformation is interpreted to have thick-skinned deformation constrained with seismicity, gravity modeling, and crystalline nappes of the Higher Himalayas and the Longmen Shan mountains (Figures 11 and 12)(4)Based on Bouguer gravity modeling and tomographic studies, the Moho is generally located at ~40-50 km depth representing full-thickness continental crust in the Himalayan foreland (Figure 13(a)), Fergana, and Tarim Basin [5, 13, 20] and 70-82 km depth representing duplicated crust in the Pamir and TP [13, 20, 21]. It is located at ~33 km depth representing thinner ~20-25 km ocean-continent transitional crust in the SFB (Figure 13(b)) along the western margin of the Indian plate [14](5)The LAB is located at ~130 to 200 km in the Himalayas and the TP (Figure 4)(6)Linear NE-SW band of seismicity to 300 km depth is located along the Hindukush with an interpretation of sinking and segmented mantle lithospheres of dual dip, with NW- and S-dipping Benioff zones in the Hindukush and Pamir [2, 6, 20](7)The Himalayan arc is the site of seismic hazard with the recorded occurrence of great earthquakes [40, 41]. Six devastating earthquakes have occurred, with <20 km hypocenter depths, along the Himalayan arc and sutures, during the last two decades. Of these, the 2005 Mw 7.6 Kashmir [31] and 2008 Mw 7.9 Wenchuan [15] earthquakes have been most devastating with >80,000 casualties each. Ground rupture of ~70 km with a vertical slip of ~7 m was mapped along the trace of the active Balakot-Bagh fault due to the 2005 Kashmir earthquake (Figure 14). Mitigation of earthquake hazards along the Himalayan arc and TP is required to avert disasters with great earthquakes(8)The margins of the TP are seismically active with limited shallow seismicity in the center of the plateau along the sutures and rigid blocks. The deformation below LVZ in the TP is inferred to be aseismic with ductile slip(9)Present-day deformation is quantified with Global Positioning Satellite (GPS) measurements in the HMS and the TP [9, 30, 42]. It shows variable slip with shallow crustal deformation and thrusting as the main mechanisms of deformation followed by extension, extrusion, and rotation of rigid blocks between colliding plates. Nanga Parbat (8126 m) exhibits 20 mm/yr westward slip [42], whereas the Karakoram, Kohistan, and Ladakh are interpreted with ~12 mm/yr SSE slip which is limited to only 5 mm/yr in the SRPP [42]. The MHT is modeled to have been locked at ~15 km depth with about 13 mm/yr ductile slip. Similarly, lower shortening rates of 3-6 mm/yr are recorded by analog and numerical models across the SFB [39, 43], consistent with GPS measurements of 5 mm/yr in the northern Kirthar Ranges [44]. Thus, variable strain accumulation depicting stick-slip deformation is inferred with potential seismic hazard in the NW Himalayas and KarakoramThe HMS is a result of a collision between the Eurasian and the Indian plates, since about 55 Ma [1]. Ongoing collision has led to the duplication of crust under the TP with the Moho at ~70-82 km depth [13, 21]Active brittle deformation is mostly confined to ~20 km depth along a narrow zone of deformation of steep topography along the frontal part of the mountain system and along sutures and faults in the internal part of the TP [5, 7, 15, 17, 31, 36]. Intermediate-depth (up to 300 km) seismicity is recorded along the Hindukush and Pamir with an interpretation of the sinking lithosphere of the Indian and the Eurasian plates [6, 20]The foreland deformation is interpreted to have a thin-skinned imbricate style of deformation in the Salt Range-Potwar Plateau (SRPP) and duplex style of deformation in the Sulaiman fold belt (SFB), constrained with industry seismic reflection profiles [18, 19, 37, 38] and seismicity [7, 39] (Figures 8–10). The hinterland deformation is interpreted to have thick-skinned deformation constrained with seismicity, gravity modeling, and crystalline nappes of the Higher Himalayas and the Longmen Shan mountains (Figures 11 and 12)Based on Bouguer gravity modeling and tomographic studies, the Moho is generally located at ~40-50 km depth representing full-thickness continental crust in the Himalayan foreland (Figure 13(a)), Fergana, and Tarim Basin [5, 13, 20] and 70-82 km depth representing duplicated crust in the Pamir and TP [13, 20, 21]. It is located at ~33 km depth representing thinner ~20-25 km ocean-continent transitional crust in the SFB (Figure 13(b)) along the western margin of the Indian plate [14]The LAB is located at ~130 to 200 km in the Himalayas and the TP (Figure 4)Linear NE-SW band of seismicity to 300 km depth is located along the Hindukush with an interpretation of sinking and segmented mantle lithospheres of dual dip, with NW- and S-dipping Benioff zones in the Hindukush and Pamir [2, 6, 20]The Himalayan arc is the site of seismic hazard with the recorded occurrence of great earthquakes [40, 41]. Six devastating earthquakes have occurred, with <20 km hypocenter depths, along the Himalayan arc and sutures, during the last two decades. Of these, the 2005 Mw 7.6 Kashmir [31] and 2008 Mw 7.9 Wenchuan [15] earthquakes have been most devastating with >80,000 casualties each. Ground rupture of ~70 km with a vertical slip of ~7 m was mapped along the trace of the active Balakot-Bagh fault due to the 2005 Kashmir earthquake (Figure 14). Mitigation of earthquake hazards along the Himalayan arc and TP is required to avert disasters with great earthquakesThe margins of the TP are seismically active with limited shallow seismicity in the center of the plateau along the sutures and rigid blocks. The deformation below LVZ in the TP is inferred to be aseismic with ductile slipPresent-day deformation is quantified with Global Positioning Satellite (GPS) measurements in the HMS and the TP [9, 30, 42]. It shows variable slip with shallow crustal deformation and thrusting as the main mechanisms of deformation followed by extension, extrusion, and rotation of rigid blocks between colliding plates. Nanga Parbat (8126 m) exhibits 20 mm/yr westward slip [42], whereas the Karakoram, Kohistan, and Ladakh are interpreted with ~12 mm/yr SSE slip which is limited to only 5 mm/yr in the SRPP [42]. The MHT is modeled to have been locked at ~15 km depth with about 13 mm/yr ductile slip. Similarly, lower shortening rates of 3-6 mm/yr are recorded by analog and numerical models across the SFB [39, 43], consistent with GPS measurements of 5 mm/yr in the northern Kirthar Ranges [44]. Thus, variable strain accumulation depicting stick-slip deformation is inferred with potential seismic hazard in the NW Himalayas and KarakoramThe Himalayan region is most heavily inhabited with risks of the highest degree of seismic hazards in the world. Figure 1 shows a seismic risk map with epicenters of deadly earthquakes. The most devastating earthquakes are located along the margins of the TP (Feng et al. [4], with sporadic occurrence in the internal parts of the system, such as the suture zones, NPHM, and the Karakoram fault.Figure 5 shows a real distribution of seismicity to ~300 km depth across the NW Himalayas, Hindukush, and Pamir [2]. Figure 5(f) shows limited seismicity at 250-300 km depth along the SW Hindukush that increases at 200-250 km depth with NW propagation (Figure 5(e)). The seismicity continues to propagate towards NW swinging towards east at 100-200 km depth along the northern Pamir with its termination in the Karakoram and Altyn fault (Figures 5(c) and 5(d)). The seismicity is significantly reduced at 50-100 km depth along the Pamir (Figure 5(b)). At shallow depth (<50 km) in the Tajik Basin, the Tien Shan, and the NW Himalayas, widespread occurrence of sporadic seismicity is recorded (Figure 5(a)). The review of seismicity shows its deepest occurrence in the Hindukush due to the sinking lithosphere with shallow crustal deformation along the NW Himalayas, the Pamir, and the TP [4, 17].Figure 6 shows a sectional view of seismicity with earthquake hypocenters plotted on NS and EW crustal profiles across the NW Himalayas. It shows underplating of the Indian plate as far north as ~36.5°N with a gentle Benioff zone on profile “A” along 70° (Figure 6). The highest concentration of earthquake hypocenters, with a steep Benioff zone on profiles “B and C,” along 71° and 72°E, below the Hindukush, is observed to ~250 km depth (Figure 6). The pattern of seismicity is revered near 71° with the south-dipping Asian lithosphere to ~160 km depth abutted with the NW-dipping Indian lithosphere, with hypocenter depths to ~275 km (Figure 6, profile C). The profile “E” along 73° shows rebound of the south-dipping Asian lithosphere to moderate dips (Figure 6) and its steepening again with profile “F” along 74° (Figure 6), implying segmentation or twisting of a sinking slab that apparently disappears along profile “G” further to the east (Figure 6). An EW profile “M” along 37° across the Tajik and the Tarim Basin shows the moderately east-dipping Asian lithosphere beneath the Hindukush and southern Pamir (Figure 6). The EW profile “L” along 36° shows a west-dipping Benioff zone related to sinking Indian mantle lithospheres with hypocenter depths to ~275 km below the Hindukush (Figure 6). Molnar and Bendick [6] have interpreted the sinking mantle lithosphere below the Hindukush with a higher rate of ~40 mm/year relative to the overlying crust. Our observations favor models of dual dip, with an inference of segmented, sinking mantle lithospheres in the Hindukush and the Pamir.In the NW Himalayas, linear seismicity is located along the eastern part of the MKT, Indus-Kohistan Seismic Zone (IKSZ) of the NW-SE Himalayan trend, and foreland parts of the SFB and along the Chaman fault, with diffused seismicity in the northern part of the Kohistan and Himalayan foreland, depicting zones of active deformation (Figures 2 and 10). Both the IKSZ and the SFB show thrusting as a mechanism of deformation, evidenced by shallow <20 km depth of earthquake hypocenters (Figures 10 and 12). The focal mechanism solution combined with surface slip and earthquake hypocenters shows left-lateral strike-slip deformation along the Chaman fault to a depth of ~22 km (Figure 10). The Chaman fault is interpreted to accommodate brittle oblique slip along the western margin of the Indian plate over the underplating Indian lithosphere beneath the Afghan block (Figure 13(b)).The lithospheric structure across the orogen has been analyzed based on gravity modeling and seismological modeling [3, 4, 13, 14, 21, 22, 45–50].The Bouguer anomalies vary from +20 to -600 mGal across the HMS and the TP (Duroy et al., 1089; [21, 47]). The lowest values are recorded along the central part of the TP (Figure 3). The near-zero Bouguer gravity values, recorded in the Himalayan foredeep in north Pakistan, descend gently to ~-200 mGal across the foreland with rapid decrease reaching -400 mGal across the MMT in north Pakistan [13, 14]. These are modeled with a full-thickness continental crust of ~40 km from the foredeep to the suture zone and duplicated crust with the Moho at ~70 km below the Kohistan arc and the Karakoram mountains in north Pakistan (Figure 13(a)). Unlike north Pakistan, the Bouguer anomalies are recorded from near zero in the foreland of the SFB to -300 in the Afghan block, along an NW-SE profile across the western margin of the Indian plate (Figure 13(b)). These are modeled with an ocean-continent transitional crust of 20-33 km below the SFB and thicker duplicated crust, with the Moho at ~60 km, below the Afghan block. On the Tibetan side, the Moho is located at ~54 km depth underneath the Qaidam Basin [45], ~40 km depth below Ferghana Valley north of the Tien Shan mountains (Figure 7), and ~82 km depth in the TP [21].Several seismic experiments have been conducted across the TP including INDEPTH [51, 52], Hi-CLIMB [24], and ANTILOPE [21, 22] to understand the seismic discontinuities and lithospheric structure. The INDEPTH-I and INDEPTH-II interpret the presence of the Indian lithosphere below central Tibet [52, 53] along with partial melting in the middle crust of central Tibet [54, 55]. Results of INDEPTH-III show decreasing crustal thickness from south to north, with a value of 70-80 km south of the middle of the Lhasa terrain to less than 60 km in the Qaidam Basin [25, 56]. This is supported by INDEPTH-IV interpreting a relatively thinner segment of the Tibetan lithosphere in central and northern Tibet over the south-dipping Asian lithosphere [57]. The Hi-CLIMB and ANTILOPE have proposed underthrusting of the Indian plate below southern Tibet [22, 24] near the Jinsha suture [21] in the TP (Figures 1 and 4). Discrete mechanical layers beneath the central TP are interpreted as LVZ at ~20 km, the Moho at ~82 km, and LAB at ~200 km depth, with wedging of the Asian lithospheric mantle between the crust and mantle of the Indian plate (Figure 4). We have revised this model of lithospheric deformation with duplexing in the crust and upper mantle, similar to the interpretation of the crustal duplex based on deep seismic reflection profiles, along the Zangbo suture in the western Himalayas [58]. The structural model proposed herein is consistent with the pattern of shallow seismicity with thrusting, strike-slip, and normal faulting [4, 5, 59] above and ductile slip below LVZ in the TP.Active deformation in the NW Himalayas and Tibet is manifested by active faults [60, 61] and seismicity [3, 5, 6, 15–17, 23, 32] with the following details.Structural features in the NW Himalayas and adjoining regions follow the NW-SE Himalayan and NE-SW Hindukush trend, with change across the Nanga Parbat Haramosh Massif (NP), that represents an active north-south oriented spur of the lithosphere of the Indian plate at the western terminus of the Himalayas (Figure 2). The NP is bounded between active Raikot fault and Diamer shear evidenced with fault scarps and hot springs towards west and Rupel shear towards east with an interpretation of a crustal pop-up, with rapid uplift of the Indian lithosphere [62]. The age dating of 01-10 Ma plutonic rocks along the length of the massif [62, 63] suggests crustal melting with implications of deeper penetration of crustal shears. The presence of hot springs aligned with faults and fault scarps in the Himalayas and Karakoram in north Pakistan represents zones of active deformation [64].The highest concentration of linear seismicity is located with a NE-SW trend along the Pamir-Hindukush Seismic Zone (PHSZ), a NW-SE Himalayan trend along the IKSZ, and the western terminus of the MKT suture representing these regions as the most active (Figure 2). The 2005 Mw 7.6 Kashmir earthquake has a hypocenter depth of 11 km along a thrust in the IKSZ with fault rupture of ~70 km [65] and ground separation of ~7 m (Figure 14). With ~87,000 casualties, it is recorded as the most devastating in the Himalayan chain, similar to the 2005 Mw 7.9 Wenchuan earthquake in eastern Tibet (Figure 11).The Himalayan foreland is represented by SRPP of a broad (>130 km) cross-sectional area and gentle topography which is generally aseismic due to the presence of evaporates along a weak décollement in the Precambrian evaporates (Figure 8). With the absence of fault scarps at the mountain front, the active deformation is interpreted to have propagated, 15 km south, along the overpressured Lilla anticline (Figures 2 and 8(a)). The Himalayan foreland deformation in India is represented by a narrow (<30 km) cross-sectional area of steeper topography with several great earthquakes due to strong friction along the décollement, reflected by active fault scarps along the MHT and MBT [17, 60, 61, 66]. Recently, an earthquake of Mw 5.6 has occurred east of the SRPP causing 40 casualties in an area where Precambrian evaporates are apparently absent. The earthquake hypocenter is located at ~10 km depth with the WNW-ESE Himalayan trend of nodal planes, westward of the Mw 08 1905 Kangra earthquake which has claimed 20,000 human lives [66]. The Mirpur earthquake indicates about likely occurrence of a great detachment earthquake in north Pakistan where evaporates are presumably absent.Figure 12 shows a regional cross section with earthquake hypocenters and structural geometry in the NW Himalayas in north Pakistan. The earthquake hypocenters are commonly located along a décollement at ~10-15 km depth in the foreland. They are at 15-25 km depth along a crustal ramp with triangle zone geometry of an active wedge in the hinterland (Figure 12). Another zone of linear seismicity, termed as the Hazara Low Seismic Zone (HLSZ) in the Hazara fold belt (HFB), to the south of IKSZ is located, with an interpretation of crustal deformation along with seismic hazard [67].The broad (>350 km across strike) SFB of gentle topography, similar to the SRPP, is located over a weak décollement at 10 and 20 km depths. The foreland part of the décollement is seismically most active in Pakistan [7, 16, 38, 44, 68, 69], supported with active fault scarps [60], along folds at the mountain front and faults in the internal part of the system (Figure 9). Active deformation in the SFB and the Chaman fault zone is represented with devastating earthquakes in the region, such as the 1935 Quetta earthquake that have claimed ~35000 casualties [70], and magnetostratigraphy [71]. Age dating by magnetostratigraphy shows that the continental Siwaliks, deposited between 700 and 50 ka, are overlain by alluvial fan deposits. The latter are tilted along the mountain front.The seismicity is generally confined to ~20 km depth, representing shallow crustal deformation in the TP. Thrusting is recognized as a key process accommodating 90% of slip along the margin of the TP in the Himalaya and the Altyn-Tagh [4, 5, 9, 65]. Its central part is generally aseismic, with the exception of limited seismicity along thrusts and normal and strike-slip faulting along laterally extruding blocks between colliding continents [9, 30]. Three major earthquakes in the last 20 years have occurred in the TP [4]. The Mw 7.8 Kunlun and Mw 6.9 Yushu earthquakes occurred with left-lateral strike-slip deformation along the Kunlun and Jinsha sutures at depths of 15 and 17 km, respectively. The Kunlun earthquake is reported to have a ground rupture of 400 km [4]. The Mw 7.9 Wenchuan earthquake has occurred from ~20 km depth along NE-SW trending thrusts in the Longmen Shan mountains (Figure 11). The Yushu and Wenchuan earthquakes have been disastrous with ~2600 and 80,000 casualties, respectively. The latter is located in the seismically active foreland part of the system with threats of future devastating earthquakes.Figure 15(a) represents a structural cross section with earthquake hypocenters across the main Himalayas. It shows a décollement descending gently northwards from ~3 km depth in the foredeep to ~15 km depth in the hinterland. The décollement is interpreted to have a ramp from 15 to 40 km below the Higher and Tethyan Himalayas and a flat ~40 km depth below the TP. The great Himalayan earthquakes, such as 1905 Mw 8 Kangra, 1930 Mw 7.1 Dhubrin, and 1934 Mw 8.3 Bihar, are located along the décollement, with an active narrow zone of steep (~30°) topography [17]. More recently, the 2015 Mw 7.8 Gorkha earthquake has occurred in the internal part of the system at ~8 km depth with ~9000 casualties in Nepal [72]. Active faults are thoroughly mapped along the mountain front and in the internal part of the system ([60]; Yeats and Thakur [61]. The Himalayan chain with a narrow zone of steep topography is recognized with threats of future great earthquakes [40, 41].We interpret the HMS and the TP as having active brittle and ductile deformation. The brittle deformation is generally confined to ~20 km depth along the décollement and LVZ in the HMS and TP based on seismic reflection interpretation [15, 18, 19] and analyses of earthquake hypocenters [2, 5, 7, 16, 17]. The ductile deformation is inferred below ~20-25 km depth, due to lack of seismicity along the lower crust and upper mantle, in the TP (Figure 15(b)).Foreland fold and thrust belts are interpreted to have imbricate systems [73], with progressive active deformation representing in-sequence and out-of-sequence deformation [74] in the external and internal part of the system. The Great Himalayan earthquakes are located along MHT representing in-sequence and out-of-sequence deformation (Figure 15(a)). The out-of-sequence deformation tends to increase the critical taper of the wedge with internal deformation in order to enable it to grow outwards with thrusting following the mechanics of deformation [75, 76]. A wedge at the critical taper can slide along its basal décollement without internal deformation with the release of strain along an earthquake. Thus, stick-slip deformation with strain accumulation and release at one or other location is interpreted. We suggest that great earthquakes along the Himalayan arc occur as a result of progressive in-sequence and out-of-sequence deformation evidenced by seismicity [2, 16, 17], magnetostratigraphy [71, 77], and field mapping [60, 61, 66, 78]. The wider zone of Himalayan foreland deformation in north Pakistan is less likely to face the threat of great earthquake as compared to the Main Himalayas in India, due to a thick layer of Precambrian evaporates over the basement [66, 79].Figure 4 shows a structural model of the HMS and the TP [21], with structural discontinuities along the (1) décollement at ~5-20 km depth, (2) LVZ at ~20-25 km, (3) Moho at 40 to 82 km, and (4) LAB at ~130-200 km depth in the Himalayas and the TP. The MHT descends steeply along a ramp below the Higher and Tethyan Himalayas, with flats at ~50 km and~82 km depths. At depth, wedging of the Tibetan mantle lithosphere in the Indian mantle lithosphere is interpreted to accommodate a space between the Moho at 82 km and LAB at 200 km, respectively.We have revised this model with flat-ramp-flat geometry of the MHT with imbrications of strata above the upper flat and duplication of the Indian lower crust between the LVZ and Moho and lithospheric mantle between the Moho and LAB (Figure 15(b)). In this model, the LVZ at ~20-25 km and Moho at ~82 km depth are proposed to act as roof and floor thrusts with ductile slip. The wedging of the Tibetan lithospheric mantle in the Indian lithospheric mantle is similar to that proposed by Xu et al. [21]. Our model suggests the presence of an anticlinal stack with duplexing of (1) lower crust and (2) lithospheric mantle of the Indian/Tibetan plates between the LVZ, Moho, and LAB. This is consistent with the interpretation of crustal-scale duplexing below the Zangbo suture in the western Himalayas [58]. In our interpretation, deformation partitioning is proposed with brittle and ductile deformation above and below LVZ, respectively.The mechanics of deformation of the TP is proposed to be similar to that of the deformation of fold and thrust belts with the presence of deformed wedge above a décollement over the undeformed basement [75, 80]. The wedge is required to maintain a critical taper between the topographic and basement slopes for thrusting. Progressive deformation occurs in the foreland with in-sequence deformation followed by that in the hinterland with out-of-sequence deformation to regain the desired taper for thrusting. The taper is influenced by the strength of the décollement, dip of the detachment, and strength of the wedge [75]. The low friction along the weak décollement leads to a wider zone of deformation of gentle topography as compared to a narrow zone of steep topography along a strong décollement [79]. The Himalayan foreland deformation is represented by a narrow zone (~30 km) of steep topography in India, as compared to a wide zone (>1000 km) of gentle topography across the TP, which is compatible with strong friction along the décollement with threats of great earthquakes along the Himalayan arc and weak friction with ductile slip along LVZ in the TP. The LVZ, Moho, and LAB are proposed to act as incompetent surfaces bounding structural layers of the lower crust and upper mantle (Figure 15(b)).The deformation of the HMS and the TP is analyzed with the following conclusions. (1)Brittle deformation to a shallow ~20-25 km depth is interpreted in the HMS and the TP, whereas ductile deformation is interpreted below LVZ at ~20-25 km and LAB at 200 km depth(2)The shallow crustal deformation is dominated by thrusting along the margins and normal/strike-slip faulting along rigid blocks above LVZ in the TP, whereas deep crustal and lithospheric deformation is interpreted with ductile slip along the Moho and LAB leading to the duplication of the crust and upper mantle as an anticlinal stack(3)Full-thickness (~40 km) continental and thin (15-27 km) oceanic/transitional crust is modeled by gravity modeling, depicting advance and early stages of continental collision along the main Himalaya and the western margin of the Indian plate, respectively(4)The seismicity of the Hindukush-Pamir region to ~300 km depth is interpreted with a sinking mantle lithosphere (north- and south-dipping), which is consistent with the early stage of continental collision, evidenced with the presence of ocean-continent transitional crust along the western margin of the Indian plate(5)Seismicity is highly concentrated along the margins of the TP, with in-sequence deformation in the foreland and out-of-sequence deformation in the hinterland(6)The mechanism of deformation of the TP is considered to be analogous to deformation of the fold and thrust belts and accretionary prisms [79] following critical wedge theory [75, 80] that suggests the presence of a décollement below a deforming wedge, similar to a wedge of snow or soil in front of a bulldozer. The wedge is interpreted to have a steep and gentle topography along zones of strong and weak friction, respectively(7)The narrow cross-sectional area of steep (~30°) topography along the HMS in India and the wider cross-sectional area of gentle topography (~1°) in TP are compatible with the presence of strong friction along the décollement in the main Himalayas and with the inference of weak friction along the Moho and LAB in the TPBrittle deformation to a shallow ~20-25 km depth is interpreted in the HMS and the TP, whereas ductile deformation is interpreted below LVZ at ~20-25 km and LAB at 200 km depthThe shallow crustal deformation is dominated by thrusting along the margins and normal/strike-slip faulting along rigid blocks above LVZ in the TP, whereas deep crustal and lithospheric deformation is interpreted with ductile slip along the Moho and LAB leading to the duplication of the crust and upper mantle as an anticlinal stackFull-thickness (~40 km) continental and thin (15-27 km) oceanic/transitional crust is modeled by gravity modeling, depicting advance and early stages of continental collision along the main Himalaya and the western margin of the Indian plate, respectivelyThe seismicity of the Hindukush-Pamir region to ~300 km depth is interpreted with a sinking mantle lithosphere (north- and south-dipping), which is consistent with the early stage of continental collision, evidenced with the presence of ocean-continent transitional crust along the western margin of the Indian plateSeismicity is highly concentrated along the margins of the TP, with in-sequence deformation in the foreland and out-of-sequence deformation in the hinterlandThe mechanism of deformation of the TP is considered to be analogous to deformation of the fold and thrust belts and accretionary prisms [79] following critical wedge theory [75, 80] that suggests the presence of a décollement below a deforming wedge, similar to a wedge of snow or soil in front of a bulldozer. The wedge is interpreted to have a steep and gentle topography along zones of strong and weak friction, respectivelyThe narrow cross-sectional area of steep (~30°) topography along the HMS in India and the wider cross-sectional area of gentle topography (~1°) in TP are compatible with the presence of strong friction along the décollement in the main Himalayas and with the inference of weak friction along the Moho and LAB in the TPThe authors declare that they have no conflicts of interest.This study was supported by Higher Education Commission Grant 20-2062, Pakistan Science Foundation Grant PSF/NSFC-Earth/KP-COMSATS-Abt(09), National Science Foundation of China Grant 41490615, and Chinese Academy of Science Grant XDA20070301.
更新日期：2021-01-08
京公网安备 11010802027423号