Seismotectonics of the Trans-Himalaya, Eastern Ladakh, India: Constraints from moment tensor solutions of local earthquake data

    Seismotectonics of the trans-Himalaya, Eastern Ladakh, India: constraints from Moment Tensor Solutions of local earthquake data Devajit Hazarika, Arpita Paul, Monika Wadhawan, Naresh Kumar, Koushik Sen, C.C. Pant PII: DOI: Reference:

S0040-1951(17)30001-X doi:10.1016/j.tecto.2017.01.001 TECTO 127373

To appear in:

Tectonophysics

Received date: Revised date: Accepted date:

1 September 2016 18 December 2016 1 January 2017

Please cite this article as: Hazarika, Devajit, Paul, Arpita, Wadhawan, Monika, Kumar, Naresh, Sen, Koushik, Pant, C.C., Seismotectonics of the trans-Himalaya, Eastern Ladakh, India: constraints from Moment Tensor Solutions of local earthquake data, Tectonophysics (2017), doi:10.1016/j.tecto.2017.01.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Seismotectonics of the trans-Himalaya, Eastern Ladakh, India: constraints

PT

from Moment Tensor Solutions of local earthquake data

RI

Devajit Hazarika1*, Arpita Paul1, Monika Wadhawan1, Naresh Kumar1, Koushik Sen1, C.C. Pant2

Wadia Institute of Himalayan Geology, 33 GMS Road, Dehradun, India

2

Kumaun University, Nanital, India

NU SC

1

*Corresponding author: [email protected]

MA

Abstract

The seismotectonic scenario of northwest part of India-Asia collision zone is studied

ED

by analyzing the local earthquake data (M~1.4-4.3) recorded by a broadband seismological network consisting of 14 stations. Focal Mechanism Solutions (FMSs) of 13 selected

CE PT

earthquakes were computed through waveform inversion of three-component broadband records. Depth distribution of the earthquakes and FMSs of local earthquakes obtained by waveform inversion reveal kinematics of the major fault zones present in eastern Ladakh. A

AC

most pronounced cluster of seismicity is observed in the Karakoram Fault (KF) zone down to a depth of ~65 km. The FMSs reveal transpressive environment with an inferred strike slip fault plane parallel to the KF. It is argued that the KF penetrates down to the lower crust and is a manifestation of active under thrusting of Indian lower crust beneath Tibet. Two clusters of microseismicity are

observed at a depth range 5-20 km at the northwestern and

southeastern fringes of the Tso Morari gneiss dome, which can be correlated to the activities along the Zildat fault and Karzok fault, respectively. The FMSs obtained for representative earthquakes show thrust fault solutions for the Karzok fault, and normal fault solutions for the Zildat fault. It is suggested that the Zildat fault is acting as a detachment, facilitating the exhumation of the Tso Morari dome. On the other hand, the Tso Morari dome is under 1

ACCEPTED MANUSCRIPT

thrusting the Karzok ophiolite at its southern margin along the Karzok fault due to gravity

PT

collapse.

MA

NU SC

Karakoram Fault Zone, Tso Morari Crystallines

RI

Keywords: Microseismicity, Seismotectonics, Trans-Himalaya, Indus Suture Zone,

1. Introduction

ED

The Cenozoic Himalayan orogeny is a consequence of northward movement and subsequent collision and under thrusting of the Indian continental plate beneath its Eurasian counterpart

CE PT

(Dewey and Bird, 1970; McKenzie and Sclater, 1971). The eastern Ladakh-Karakoram zone, the northwest part of the Trans-Himalayan belt, bears signature of this collisional process in the form of suture zones, exhumed blocks that underwent deeper subduction and also intra-

AC

continental fault zones. This region, extending towards north into the Tibetan plateau, has been studied extensively in the recent past; the major lithospheric features including (a) presence and extent of mid-crustal low velocity zone, (b) depth of the Mohorovicic (Moho) discontinuity, (c) northern extent of the under thrusting Indian plate and (d) depth of the Lithosphere-Asthenosphere Boundary (LAB) are well constrained with the help of numerous passive seismological studies (Kind et al., 2002; Wittlinger et al., 2004; Schulte-Pelkum et al., 2005; Kumar et al., 2006a, 2006b; Rai et al., 2006; Nábělek et al., 2009; Zhao et al., 2010; Zhao et al., 2011; Hazarika et al., 2014; Gilligan et al., 2015). However, as the TransHimalaya is seismically less active, the crustal structures or present day kinematics of major 2

ACCEPTED MANUSCRIPT

fault zones of the region remain poorly understood. Moderate magnitude earthquakes in the

PT

region are rarely reported by global seismological networks. The present work addresses this issue with the help of microearthquake network data recorded during 2009-2012 by 14

RI

broadband seismic stations in the eastern Ladakh region. One of the important aspects of

NU SC

application of microseismic monitoring is to determine precise locations of the events as well as source mechanisms of the local earthquakes. Seismic Moment Tensors (MTs) provide key information of source mechanisms which are computed by waveform inversion technique.

MA

This technique characterizes an earthquake source with the estimation of source parameters like strike, dip, rake, seismic moment and moment magnitude. In this study, MT solutions are

ED

obtained for 13 selected earthquakes, with magnitude Mw 3.0-4.3, and then the solutions are corroborated with the geological observations to understand the seismotectonics of the

CE PT

region. 2. Geo-tectonic Setting

The eastern Ladakh region encompasses all the major tectonic units of the Trans-

AC

Himalaya (Fig. 1). The major tectonic units of the study area from SW to NE across the regional strike of the Indus Suture Zone (ISZ) are the Tso-Morari Crystalline Complex (TMC), The Zildat Ophiolitic Mélange (ZOM), the Nidar Ophiolitic Complex (NOC), the Indus Group of foreland sediments or the Indus Molasse, the Ladakh magmatic arc and various leucogranites, migmatites and metamorphics of the Karakoram Terrane. The Palaeozoic TMC is part of the deeply subducted and exhumed northern margin of the Indian continental plate and consists of ortho- and para- gneisses which bear metabasic enclaves of high pressure (HP eclogites) to ultra-high pressure (UHP coesite) metamorphic grades 3

ACCEPTED MANUSCRIPT

(Mukherjee and Sachan, 2004). The TMC is bounded by two oppositely dipping faults in its

PT

SW and NE boundaries. To the SW of the TMC lies the southerly dipping Karzok fault (Fig. 1), which separates the TMC from the Mata unit that consists of Karzok Ophiolites and

RI

Precambrian to Palaeozoic metasediments and a Palaeozoic granite, named Mata Granite (De

NU SC

Sigoyer et al., 2004; Epard and Steck, 2008). The NE boundary of the TMC is characterized by the northerly dipping Zildat Fault (ZF; Fig. 1a), which demarcates the contact between the TMC and the meta-greywackes and exotic carbonate blocks of the ZOM (De Sigoyer et al.,

MA

2004; Epard and Steck, 2008; Sen et al., 2013). The ZOM is followed by ~8 km continuous crustal and mantle section of an ophiolite of Tethyan oceanic origin, called the Nidar

ED

Ophiolitic Complex (NOC) (Mahéo et al., 2004; Ahmed et al., 2008). The foreland sediments consisting of siltstone, mudstone, carbonates and conglomerates of Palaeocene to Miocene

CE PT

make up the Indus Molasse (Searle et al., 1990; Sinclair and Jaffey, 2001). The southerly dipping Indus Thrust or the Great Counter Thrust separates the Indus Molasse from the Ladakh Magmatic Arc or the Ladakh Batholith. The Ladakh Magmatic Arc was formed due

AC

to subduction of the Neo-Tethys oceanic plate, prior to the Himalaya orogeny, during Early Cretaceous to Late Eocene (Honegger et al., 1982; Scärer et al., 1984; Weinberg and Dunlap, 2000). The Ladakh Batholith is followed up north by back-arc volcanics of felsic nature, called as the Khardung Volcanics, which are of Palaeocene to Eocene ages (Upadhyay et al., 1999; Dunlap and Wysoczanski, 2002; Bhutani et al., 2004). The ~700 km long dextral strike-slip Karakoram Fault Zone (KFZ) is suggested to be an important facilitator of eastward flow of the Tibetan crust, owing to India-Eurasia convergence (Peltzer and Tapponnier, 1988; Armijo et al., 1989; Tapponnier et al., 1996). In the present study area, the KFZ is divided into two strands, the Tangtse strand to the SW and the Pangong strand to the 4

ACCEPTED MANUSCRIPT

NE as shown in Fig. 1a (Searle et al., 1998). The zone lying in between these two strands is

PT

called the Pangong Transpression Zone.

RI

3. Seismological network and data

The Wadia Institute of Himalayan Geology (WIHG), Dehradun established 10 broad

NU SC

band seismological stations in the eastern part of Ladakh-Karakoram zone for passive seismological study, which was under operation during the period 2009-2012 (Fig.1). The seismological stations covered major geotectonic units of the Trans Himalaya covering Tso-

MA

Morari crystalline complex, the Indus Suture Zone, Niddar Ophilitic Complex, Ladakh Batholith and Karakoram fault zone (Hazarika et al., 2014). Due to limited accessibility and

ED

tough terrain condition, the distribution of the seismological stations were restricted to a nearly linear profile. Four seismological stations of Kinnaur network (Kumar et al., 2014),

CE PT

Himachal Pradesh (KAZA, LOSR, MUDH and HURL; Fig. 1a) are also used. These stations facilitate in achieving azimuthal coverage to the south. The earthquakes of M > 3.0 were well recorded by these stations. All these seismological stations were equipped with Trillium 240

AC

seismometer with a flat velocity response for the frequency range of 0.004–35 Hz. Taurus data logger with dynamic range>138 db (Make M/S Nanometrics, Canada) and 40 GB storage capacity hard disk were used. Global Positioning System (GPS) receivers were used in the data logger for time synchronization. The digital sampling rate of the equipments of Ladakh and Kinnaur networks were 20 and 100 samples per second (SPS) respectively. The seismological stations recorded total 317 local earthquakes occurred in and around NW Himalaya, out of which 148 best located earthquakes in the study region encompassing Tso-Morari Crystalline Complex, Ladakh batholith and Karakoram fault zone, have been selected. The selection of 148 earthquakes is done based on minimum errors in 5

ACCEPTED MANUSCRIPT

estimation of hypocentral parameters and high signal to noise ratio. The magnitude range of

PT

the selected earthquakes is 1.4 - 4.3. 4. Methodology

RI

Estimation of hypocentral parameters

NU SC

Total 317 local earthquake data have been analyzed in SEISAN software (Havskov and Ottemoeller, 2001) to estimate the arrival times of different seismic phases. Initially, the hypocentral parameters of the earthquakes are computed using the HYPO71 program of Lee

MA

and Lahr (1975). As the seismological profile is almost linear with limited azimuthal coverage, the errors in estimation of hypocentral parameters are higher for earthquakes

ED

located much away from the network. Consequently, we limited the investigation within a rectangular area (~100 km x 200 km) as shown in Fig. 1. Total 148 earthquakes are relocated

CE PT

within the rectangular area with minimum location errors; epicenter and depth within 3.5 km and root mean square (RMS) error in origin time < 0.25 s. The computation process requires one-dimensional (1D) velocity model of the region to the depth extent down to the maximum

AC

focal depth of earthquake data used. The 1D velocity model for 80 km thick lithosphere is constructed from combined information of published study of Kumar et al. (2009) for the crustal part of adjoining SW zone of NW Himalaya and Hazarika et al. (2014) for Moho and uppermost mantle of present study region (supplementary Table S1). We further applied hypoDD method (Waldhauser & Ellsworth, 2000; Waldhauser 2001) to relocate the earthquakes. This method uses the absolute and relative times to improve the source locations of a couple or cluster of earthquakes. The method is highly useful for location precision using relative times of two close source events where the geological setup is variable on local scale and no precise velocity model is available. It is based on the assumption that wave 6

ACCEPTED MANUSCRIPT

propagation takes place nearly along similar subsurface path and relative times at recording

PT

stations can be used for improving locations. In a closely spaced cluster of events, the final earthquake locations are computed based on relative to cluster centroid. Therefore, the

RI

precise locations in a few clusters are more useful for identification of seismogenic faults

NU SC

compared to individually located hypocentres. After relocation (Fig. S2), the errors in estimation of epicenter locations and depth are significantly reduced to < 2.5 km with RMS error in origin time < 0.09 s.

MA

Although the relocated hypocentres are scattered but we are able to extract 10 clusters of hypocentres consisting 148 events with appropriate cluster centroid, which contain one

ED

major cluster, two medium size and remaining small clusters (Fig. S2). A major cluster of events in the northern part close to the KF are aligned nearly in east-west direction. These

CE PT

epicentres are not aligned to the strike direction of the KF but branched into small seismogenesis zones and may be related with hidden sub-surface tectonic features of localised extent. Other

few relocated events to the southeast of the KF are aligned along the strike of this major

AC

tectonic feature suggesting that this part of the fault is seismogenically active. Two other major clusters are observed, one to the north near CHUM station in the ISZ while the other to the south in the TMC. The northern cluster may be correlated with the ZF, and the other to the south with the Karzok Fault. In the depth section, these three main clusters are clearly observed in the upper ~20 km crust (Fig. 1b). The lower crustal part produced a few events in the TMC zone to the southern part of the study region. Moment Tensor solution Moment Tensor (MT) solution is one of the most powerful tools to study source mechanisms of smaller earthquakes on regional scale as such information is usually not 7

ACCEPTED MANUSCRIPT

reported by global seismological networks (e.g. United States Geological Survey (USGS) and

PT

International Seismological Center (ISC) etc.) for smaller earthquakes. Information of smaller earthquakes can be obtained only through data recorded by local seismological stations. We

RI

computed MT solutions from waveform inversion of 13 local earthquakes (Table 1) selected

NU SC

based on high signal to noise ratio, clear record of different phase arrivals and events with comparatively higher magnitude range (Mw 3.0-4.3) and epicentral distance <100 km. Use of events with smaller epicentral distance help to reduce dependency on the crustal velocity

MA

model (Zahradnik et al., 2008).

The computer software ISOLA (www.mathworks.com/products/MATLAB) is used

ED

for waveform inversion (Sokos and Zahradnik, 2008). The inversion scheme implemented in ISOLA follows the iterative deconvolution method of Kikuchi and Kanamori (1991)

CE PT

modified for regional distances by Zahradnik et al. (2005) for waveform inversion of local earthquakes. Waveforms are related to the source mechanism and properties of propagating medium primarily by MT and the Green’s function. Complete waveforms of all the three

AC

components are used for waveform inversion without selecting any specific phase. Full wave elastodynamic Green’s functions have been computed for a given earth structure, source position and station location following the discrete wavenumber method (Bouchon, 2003). Local 1-D crustal velocity model from previous study (Kumar et al. 2009, Hazarika et al., 2014) is used for Green’s function calculation. Then the MT inversion is carried out which is a linear inverse problem and solved by least squire method. The MT inversion carry out a grid search over a set of trial source position and time shifts for obtaining optimal centroid position, time and corresponding moment tensor with minimum residual errors. The residual errors are obtained from comparison of synthetic and observed waveforms. The minimization 8

ACCEPTED MANUSCRIPT

of residual error correspond to maximizing the correlation between observed and synthetic

PT

seismograms. The source horizontal location is kept fixed to the epicenter during inversion. The matching between the observed and synthetic data is quantified by variance reduction:

RI

Varraed = 1-E/O, where E=  (Oi-Si)2 and O =  (Oi)2, where O and S stand for observed and

NU SC

synthetic data (Zahradnik et al.,2005). The MT solution has deviatoric and volumetric part. We focus on the deviatoric inversion. Deviatoric tensor decomposition approach is applied to separate the double-couple (DC) part and the compensated linear vector dipole (CLVD) part,

MA

which is non-double couple and both have relative sizes (1-2n) and (2n), respectively, where 1, n-1, and -n are the normalized MT eigenvalues (Zahradník et al., 2008). The DC

ED

percentage is estimated by the equation DC% = 100*(1-2n) following Zahradník et al., (2008). Prior to inversion, the waveform data are preprocessed by correcting for instrument

CE PT

response, filtering to a low frequency range, baseline correction, mean and trend removal. Usually the lowest available frequency range with a high signal/noise ratio is chosen for inversion to reduce the dependence of modeling procedure on precise crustal structure. The

AC

velocity records are integrated once to get displacement record. To portray the inversion process, analysis of an earthquake of Magnitude 4.2 (event data 04/10/2010, Origin Time: 03h 40m 27.5s, event no.: 5 in Table 1) is illustrated here. Waveform data of 8 stations were pre-processed and band pass filtered within frequency band 0.07-0.15 Hz. The hypocentral depth of the earthquake was 12.4 km. The best MT solution is grid searched considering five trial depth within the range 11-14 km in step of 1 km and temporal grid search between −3 and +3 seconds with respect to origin time keeping the source horizontal location fixed to the epicenter. The comparison of synthetic and observed seismogram for best fit solution is shown in Fig. 2. The best fit solution is judged based on 9

ACCEPTED MANUSCRIPT

largest value of correlation, DC% and variance reduction. Source depth vs time shift

PT

correlation plot is shown in Fig. 3, with maximum correlation normalized to unity. The acceptable solutions is chosen based on are largest normalized correlation value. Thus, the

RI

best fit solution is observed at 12 km depth with a time shift of ~0.35 s which is shown by red

NU SC

beach ball (Fig. 3). The variation of correlation values with respect to various trial time shifts and trial depths are shown in Fig. 4. The DC% percentage is 95.8 with highest correlation value (Fig. 5). The results corresponding to the best waveform fit shows 2.51E+15 Nm

MA

seismic moment which corresponds to Mw 4.2. The strike, dip and rake of the inferred fault plane is 341o, 26o and –94o, respectively. The FMS obtained through waveform inversion

ED

produces two orthogonal nodal planes. The fault plane, out of the two nodal planes, is inferred judging the strike and dip of the nodal planes with that of the geologic structure and

CE PT

hypocenter section. Following the similar way earthquake source parameters for rest of the earthquakes have been obtained. 5. Results and discussion

AC

We used 148 earthquakes of magnitude ranging from 1.4 to 4.3 at various depths, in the Trans-Himalaya spreading from the Karakoram terrane in the northeast to the Tso Morari region in the southwest (Fig. 1a). FMS for 13 earthquakes (M 3.0-4.3) from different tectonic domains were obtained applying waveform inversion technique which help us to envisage the crustal structures related to various litho-tectonic units of the present study area (Fig. 1a). The results are summarised in Table 1, and beach ball presentations of the FMSs are illustrated in Fig. 1. The double couple percentage and average variance reduction are two important parameters used to judge the quality of results. These values are listed in Table-1. The synthetic and observed waveform fits for accepted solutions and corresponding depth-time 10

ACCEPTED MANUSCRIPT

grid-searches are shown in supplementary figure (Fig. S1). The events that show rapid

PT

variation of the solution for minor changes in input parameters, low small DC component (<50%) and major waveform misfit are discarded. The fault planes are inferred out of two

RI

nodal planes based on existing geological information of the region. The parameters (e.g.

NU SC

strike, dip and rake) of inferred fault planes are listed in Table 1, and are used to characterize the existing geological faults. However, as we consider an earthquake as point source, one single fault plane does not necessarily represent the entire geological structure. With the

MA

limited data set, it is attempted to characterize the existing faults at depth. All earthquakes are projected on a vertical plane along A-B (Fig. 1b) and the FMSs are shown. The AB section is

ED

orthogonal to the NW-SE tectonic features of the Ladakh-Karakoram zone (Fig. 1a and b). In this section, we discuss the results as well as significances of our study in terms of

CE PT

seismotectonics of the trans-Himalaya from Ladakh-Karakoram region. Present day crustal architecture of the Trans-Himalaya, based on our study and existing geological knowledge, is presented in 3D block diagram (Fig. 6).

AC

The Karakoram Fault Zone

Seismicity pattern in the KFZ shows two distinct clusters at around Spangmik (SPAN) and Durbuk (DRBK) stations (Fig. 1a). Fault plane solutions of four earthquakes close to these stations (event Nos. 12 and 13 near DRBK, and 6 and 7 near SPAN) show thrust mechanisms with significant dextral strike-slip movements. Inferred fault planes suggest NW-SE orientation of the fault planes with steep dip angles 77o -86o (Table 1, Fig. 1). The double couple percentage of these events are within the range 79-93% (Table 1, Fig. S1). The strike and dip are in conformity with the KF obtained from geological observations (Rutter et al., 2007; Jain et al., 2008; Sen et al., 2014). Geological studies carried out in the 11

ACCEPTED MANUSCRIPT

past have also shown that the Pangong transpressional zone (Fig. 1), bounded by the Tangtse

PT

and Muglib strands of the KF, has exhumed by a combination of thrusting and strike-slip movements (Searle et al., 1998; Weinberg et al., 2000; Mukherjee et al., 2012; Sen et al.,

RI

2014). Microstructural investigation carried out by Rutter et al. (2007) on the calc-mylonites

NU SC

of the Muglib Strand reveals cataclastic deformation and presence of fault gauges that indicate shallow subsurface brittle deformation and recent activity along the Muglib Strand of the KF. The slicken lines, formed due to fault movement, also indicate a combination of

MA

thrusting and strike-slip motions along the Tangtse Strand (Sen et al., 2014). Based on these geological information, it can be inferred that the clusters of seismicity we observed in the

ED

Karakoram terrane in Durbuk and Spangmik are related to dextral transpression movement of the KF, which is facilitating the rapid exhumation of the Pangong transpression zone.

CE PT

The depth section also shows that the seismicity is evenly distributed and forms a cluster from shallow upper crust to a depth of ~30 km, followed by isolated events at greater depth, reaching to a maximum depth of ~65 km (Fig.1.b). Fault planes of two earthquakes at

AC

the middle crust (event No. 11 at ~21 km depth) and one event at the lower crust (event No. 4 at ~66 km depth) respectively, also strike parallel to the regional trend of the KF. The event No. 4 (Mw 4.1) was reported by USGS at a depth of 67 km. However, depth of this earthquake is observed to be ~66 km with help of relocation of the earthquake using hypoDD method as well as centroid depth estimated by moment tensor solution. As this earthquake was away from our network, special care was taken during waveform inversion. This earthquake was recorded by 5 seismological stations of the Ladakh network. The waveforms of DRBK, PHOB and SPAN were used for waveform inversion. The inversions have been performed for several trial depths (Fig. S1d). The best fit solution is obtained at ~66 km. The 12

ACCEPTED MANUSCRIPT

parameters of the inferred fault plane are: strike dip andrake. The inferred

shows thrust mechanism with dextral strike-slip component.

PT

nodal plane of the event No. 1, Mw 4.1, depth 12.7 km also strikes parallel to the KF and

RI

The subsurface extent of the KF is not well constrained. There are geological

NU SC

observations which argue that metamorphism and magmatism in a ductile regime is synchronous with activation of the KF and hence the KF is envisaged as a crustal or lithospheric-scale fault (Lacassin et al., 2004; Valli et al., 2008; Weinberg and Mark, 2008;

MA

Weinberg et al., 2009). Klemprer et al. (2013) obtained a high He3/He4 ratios from hot springs of the Karakoram zone and inferred that the KF may have pierced the tectonically active

ED

Tibetan mantle. As our study shows continuous occurrence of seismicity in the Karakoram zone down to a depth of ~30 km and further occurrence of few earthquakes at ~66 km depth;

CE PT

we envisage that the KF extends to the Indian lower crust (Fig. 1; Fig. 6). Two clusters of seismicity are observed to the north of Phobrang and Pangong Tso Lake (Fig.1a); one trending north-south in the eastern side of the Shyok River while the other trending ENE-

AC

WSW to the northeast of Phobrang. These earthquakes may be related to the intersection of KF with the Longmu-Gozha fault (van Buer et al., 2015). The Ladakh Batholith and Indus Foreland Sedimentary Basin The Ladakh Batholith is almost free from seismicity and we detected only a few shallower earthquakes of lower magnitude. The Indus Molasse has a tectonic contact with the Ladakh Batholith as it overrides the granites of this magmatic arc along the Indus Thrust (Fig. 1a), which has a strike parallel to that of the suture zone (NW-SE). We detected and analysed one shallow focus earthquake (No. 10, Mw 3.0; depth ~8 km) near Upshi at the contact zone 13

ACCEPTED MANUSCRIPT

between the Indus Molasse and the Ladakh Batholith. This area has strath terraces where

PT

Quaternary deposits override the bedrock and bears significant Quaternary deformation. We infer that this earthquake is related to the deformation along the Indus thrust as the fault plane

RI

solution suggests thrust movement and inferred fault plane strikes parallel to the strike of the

NU SC

Indus thrust (Fig. 1a; Table 1). The Indus Suture Zone and Tso Morari dome

Two distinct groups of seismic events are observed at the northwest (near CHUM

MA

station) and at the southeast fringes of the TMC (near TSMR station), primarily at depth range ~5-20 km. These shallow focus earthquakes may be correlated to the activities along

ED

the ZF to the north and Karzok fault to the south of the TMC. We obtained FMSs for two earthquakes (event Nos. 5 and 9) in the northern fringe and for two earthquakes (event Nos. 2

CE PT

and 3) in the SE fringe of the TMC. The event 5 (Mw 4.2) and 9 (Mw 3.6) occurred near the Zildat fault at 12.6 and 13.7 km depth, respectively. The FMSs show normal faulting (Fig 1a, Table 1). The inferred fault planes dip towards northeast and north with dip angles 26o and

AC

57o, respectively (Table 1). In contrast, the event no. 8 located north of the ZF beneath the NOC at 13.4 km depth, shows thrust mechanism with NE dipping (~54o) fault plane striking N270oW. Source mechanisms of two shallower earthquakes (Event Nos. 2, 3 with focal depth 9.9 and 9.6 km, respectively) in the SE fringe of the TMC characterize thrust faulting at the shallower part of the Karzok Ophiolitic zone. Based on local geology, the fault planes are inferred to be south and southwest dipping with dip angles ~52o and 53o respectively. It is noted that both the Zildat and Karzok faults bind the TMC and basically follow boundary of the elliptical gneiss dome. Hence the strikes of these faults are different, and the fault plane

14

ACCEPTED MANUSCRIPT

solutions obtained from these two fault zones also do not show same strike, though the

PT

kinematics and dip direction remain same.

RI

The ZF is characterised as a detachment fault that marks the northeastern contact of the TMC with the ophiolites and mélange rocks of the Indus Suture Zone. This region is

NU SC

characterised by two hot springs, namely the Chumathang (near CHUM station, Fig. 1) and the Puga hot springs. Magnetotelluric (MT) study carried out in the past revealed presence of deep seated hot water reservoir in Chumathang, and presence of an intrusive body at ~7 km

MA

depth in the Puga valley as possible heat source for these hot springs (Gupta et al., 1983). Subsequent MT investigations have detected shallow and mid-crustal low resistivity zones,

ED

inferred as partial melts or hydrothermal fluids (Harinarayana et al., 2006; Aziz and Harinarayana, 2007) (Fig. 6). Stable isotope analysis carried out by Sen et al. (2013) suggests

CE PT

decompression melting and/or metamorphic devolatization of the TMC due to its continued exhumation as a possible reason for generation of hydrothermal fluids and/or partial melts. It is believed that fluids present in the subsurface change rheological properties of rocks and

AC

play a major role in triggering shallow earthquakes in seismogenic zones (Sibson, 1992; Byerlee, 1993; Cox, 2005; Unsworth et al., 2005). The ZF or Zildat detachment fault is a high strained zone (Sen et al., 2013) and movement along this detachment facilitates the exhumation of the TMC (De Sigoyer et al., 2004; Epard and Steck, 2008). It can be inferred that the activity along the Zildat detachment facilitated by presence of fluids produced by metamorphic devolatization and/or isothermal decompression of the TMC. The FMSs (event nos. 5 and 9) obtained for two earthquakes at the Zildat detachment show normal faulting with minor strike-slip component in one solution (event no. 9), which supports the above inference. The FMS of one earthquake at Chumathang (event no. 8, Fig.1a) shows 15

ACCEPTED MANUSCRIPT

dominantly thrust faulting with strike-slip component. The strike of one nodal plane is

PT

comparable with the Indus Thrust where the Indus Molasse overrides the Ladakh Batholith. However, we do not rule out the probable contribution of crustal fluid for this particular

RI

seismic event.

NU SC

Detailed structural analysis and geological mapping carried out by De Sigoyer et al. (2004) and Epard and Steck (2008) show presence of the Zildat and Karzok faults at the north-eastern and south-western margin of the TMC. Both these workers concur that activity

MA

along these two faults facilitates exhumation of the TMC. Our study also reveals seismicity related to extensional movement in the Zildat fault zone, which is in agreement with the

ED

tectonic models proposed by these two groups of workers. However, fault plane solutions obtained from two earthquakes from the seismicity cluster observed in the south-western

CE PT

margin of the TMC show thrust faulting (Event Nos. 2 and 3; Fig. 1). In contrast to extensional movement along the ZF, the Karzok fault experiences thrusting. Possibly the TMC is undergoing ‘orogenic collapse’ or ‘gravity collapse’ to reactivate the detachment to

AC

develop thrust mechanism.

A few earthquakes are observed at a depth range of ~30-50 km beneath the TMC (Fig. 1b, Fig. 6). Receiver Function (RF) analysis carried out by Hazarika et al. (2014) reveals a low velocity zone, supposedly caused by presence of partial melts, at 15-40 km depth in the Trans-Himalaya. It is likely that these earthquakes at the lower parts of the mid-crustal low velocity zone are caused by competency contrast between the melt-rich Indian middle crust and the relatively drier and partially eclogitized Indian lower crust detected at ~47-50 km depth (Hazarika et al., 2014) (Fig. 6). Relative movement between these two crustal layers and delamination of the Indian lower crust may be a triggering factor for these earthquakes. 16

ACCEPTED MANUSCRIPT

6. Conclusion

PT

Based on seismicity and fault plane solutions (FMSs) of local earthquakes occurred in the Eastern Ladakh-Karakoram zone, following conclusions are made:

RI

(1) Pronounced cluster of seismicity is observed in the Karakoram Fault (KF) Zone down to

NU SC

a depth of ~66 km.

(2) The FMSs of earthquakes at different depths beneath the KF reveal transpressive environment with the strike of inferred fault plane parallel to the KF, which indicates that

MA

the KF penetrates down to the lower crust.

(3) Two clusters of shallow seismicity (at depth ~5-20 km) observed at the northwestern and

ED

southeastern fringes of the Tso Morari gneiss dome are correlated to the activities along the Zildat fault and Karzok fault, respectively.

CE PT

(4) The FMSs estimated for representative earthquakes at the Zildat fault zone show normal faulting supporting the inference that the fault is acting as a detachment plane, facilitating exhumation of the Tso Morari dome.

AC

(5) The FMSs estimated for earthquakes at the Karzok fault zone show dominantly thrust faulting. A plausible interpretation of these thrust mechanisms is that the Tso Morari dome is undergoing gravity collapse at final stage of exhumation causing reactivation of existing detachment as a thrust.

17

ACCEPTED MANUSCRIPT

Acknowledgement

PT

The authors are thankful to the Director, Wadia Institute of Himalayan Geology (WIHG), Dehradun, India, for providing support and encouragement to carry out this work. The

RI

Wildlife Protection Department, Jammu and Kashmir is sincerely acknowledged for giving

NU SC

permission to carry out the field work in Ladakh. The logistic support provided by Dr C.P. Dorjey in Leh is sincerely acknowledged. The necessary funding for this work was provided

AC

CE PT

ED

MA

by the WIHG.

18

ACCEPTED MANUSCRIPT

References:

PT

Ahmed, T., Tanaka, T., Sachan, H.K., Asahara, Y., Islam, R., Khanna, P.P., 2008. Geochemical and isotopic constraints on the age and origin of the Nidar Ophiolitic

NU SC

Indus suture zone. Tectonophysics 451, 206-224.

RI

Complex, Ladakh, India: Implication for the Neo-Tethyan subduction along the

Armijo, R., Tapponnier, P., Han, T., 1989. Late Cenozoic right-lateral strike-slip faulting in southern Tibet. J. Geophys. Res. 94, 2787– 2838.

MA

Azeez, K.K., Harinarayana, T., 2007. Magnetotelluric evidence of potential geothermal resource in Puga, Ladakh, NW Himalaya, Curr. Sci. 93,323–329. 40

Ar-39Ar dating of volcanic rocks of the

ED

Bhutani, R., Pande, K., Venkatesan, T. R., 2009.

Shyok Suture Zone in north-west trans-Himalaya: Implications for the post-collision

CE PT

evolution of the Shyok Suture Zone. J. Asian Earth Sci. 34, 168-177. Bouchon, M., 2003. A review of the discrete wavenumber method, Pure Appl. Geophys. 160, 445–465.

AC

Byerlee, J., 1993. Model for episodic flow of high pressure water in fault zones beforeearthquakes. Geology 21, 303–306. Cox, S.F., 2005. Coupling between deformation, fluid pressures, and fluid flow in ore producing hydrothermal systems at depth in the crust. In: Hedenquist, J.W., Thompson,J.F.H., Goldfarb, R.R., Richards, J.P. (Eds.), Economic Geology, 100th Anniversary Volume. Econ Geol. pp. 39–76. De Sigoyer, J., Gulliot, S., Dick, P., 2004. Exhumation of ultrahigh-pressure Tso Morari unit in eastern Ladakh (NW Himalaya): a case study. Tectonics 23, TC3003 doi:10.1029/2002TC001492. 19

ACCEPTED MANUSCRIPT

Dewey, J.F., Bird, J.M., 1970. Mountain belts and the new global tectonics. J. Geophys. Res.

PT

75, 2625–2647. Dunlap, W.J., Wysoczanski, R., 2002. Thermal evidence for early Cretaceous metamorphism

NU SC

Asian Earth Sci. 20, 481–490.

RI

in the Shyok suture zone and age of the Khardung volcanic rocks, Ladakh, India. J.

Epard, J. L, Steck, A., 2008. Structural development of the Tso Morari ultra-high pressure nappe of the Ladakh Himalaya. Tectonophysics 451, 242-264.

MA

Gilligan, A., Priestley, K. F., Roecker, S. W., Levin, V., Rai, S. S., 2015. The crustal structures of Western Himalayas and Tibet. J. Geophys. Res. 120, 3946-3964.

ED

Gupta, M. L., Sharma, S. R., Drolia, R. K., Singh, S. B., 1983. Subsurface thermal conditions

59.

CE PT

of Puga Valley hydrothermal field, Himalaya, India. Journal of Geophysics 54, 51-

Harinarayana, T. et al. 2006. Exploration of geothermalstructure in Puga geothermal field, Ladakh Himalayas, India by magnetotelluric studies, J. appl. Geophys.,58(4), 280–

AC

295.

Havskov, J. and Ottemoeller, L., 2001. Seisan: The earthquake analysis software user manual. Institute of Solid Earth Physics, University of Bergen, Norway, pp-277. Hazarika, D., Sen, K., Kumar, N., 2014. Characterizing the intracrustal low velocity zone beneath northwest India-Asia collision zone. Geophys. J. Int. 199, 1338-1353. Honegger, K. et al. 1982. Magmatism andmetamorphism in the Ladakh Himalayas, the Indus Tsangpo suture zone. Earth Planet SciLett 60, 253–292.

20

ACCEPTED MANUSCRIPT

Jain, A.K., Singh, S., 2008. Tectonics of the southern Asian Plate margin along the

PT

Karakoram Shear Zone: constraints from field observations and U-Pb SHRIMP ages. Tectonophysics 451, 186-205.

RI

Kikuchi, M., Kanamori, H., 1991. Inversion of complex body waves-III., Bull. Seismol. Soc.

NU SC

Am. 81, 2335–2350.

Kind, R. et al., 2002. Seismic images of crust and upper mantle beneath Tibet: evidence for Eurasian plate subduction, Science 298, 1219–1221.

Klemperer, S. L., Kennedy, B. M., Sastry, S. R., Makovsky, Y., Harinarayana, T., Leech, M.

Planet. Sci. Lett. 366, 59-70.

MA

L., 2013. Mantle fluids in the Karakoram fault: Helium isotope evidence. Earth

ED

Kumar N., Sharma, J., Arora, B. R., and Mukhopadhyay, S., 2009. Seismotectonics model of

CE PT

the Kangra-Chamba sector of Northwestern Himalaya: constraints from joint hypocenter determination and focal mechanism. Bull. Seismol. Soc. Am 99, 95-109. Kumar N., Mate, S., and S. Mukhopadhyay, 2014. Estimation of Qp and Qs of Kinnaur

AC

Himalaya. J. Seismology,18(1), 47-59, doi:10.1007/s10950-013-9399-7. Kumar, P. et al. 2006a. The lithosphere-asthenosphere boundary in the Tien ShanKarakoram region from S receiver function: Evidence for continental subduction. Geophys. Res. Lett. 32, L07305. Kumar, P. et al. 2006b. Imaging the colliding Indian and Asian lithospheric plates beneath Tibet. J. Geophys. Res. 111, B06308. Lacassin, R.et al. 2004. Large-scale geometry, offset and kinematic evolution of the Karakorum fault, Tibet. Earth Planet. Sci. Lett. 219, 255–269.

21

ACCEPTED MANUSCRIPT

Lee, W. H. K., and Lahr, J. C. ,1975. HYPO71 (revised): A computer program for determin-

Geological Survey Open-File Report 75-311, 116 pp.

PT

inghypocenter, magnitude and first motion pattern of local earthquakes. U.S.

RI

Mahéo, G., Bertrand, H., Guillot, S., Villa, I.M., Keller, F., Capiez, P., 2004. The south

NU SC

Ladakh ophiolites (NW Himalaya, India): An intra oceanic tholeiitic origin with implication for the closure of the Neo-Tethys.Chem. Geol. 203, 273–303. McKenzie, D., Sclater, J.G., 1971. The evolution of the Indian Ocean since the Late

MA

Cretaceous. Geophys. J. Int. 24, 437–528.

Mukherjee, B. K., Sachan, H. K., 2004. Discovery of coesite from Indian Himalaya: A record

CE PT

1358–1361.

ED

of ultra-high pressure metamorphism in Indian continental crust. Curr. Sci. 81,

Mukherjee, B. K., Sen, K., Sachan, H.K., Paul, S.K., 2012. Exhumation history of the Karakoram fault zone mylonites: new constraints from microstructures, fluid inclusions and 40Ar/39Ar analyses. Lithosphere 4, 230-241. DOI 10.1130/L163.1

AC

Nábělek, J.et al.2009.Underplating in the Himalaya-Tibet collision zone revealed by HiCLIMB experiment. Science 325, 1371-1374. Peltzer, G., Tapponnier, P., 1988. Formation and evolution of strike-slip faults, rifts, and basins during the India-Asia collision: An experimental approach. J. Geophys. Res. 93: doi: 10.1029/88JB03726. issn: 0148-0227. Rutter, E. H., Faulkner, D. R., Brodie, K. H., Phillips, R. J., Searle, M. P., 2007. Rock deformation processes in the Karakoram fault zone, Eastern Karakoram, Ladakh, NW India. J. Struct. Geol. 29, 1315-1326. 22

ACCEPTED MANUSCRIPT

Rai, S. S., K. Priestley, Gaur, V. K., Mitra, S., Singh, M. P., and Searle, M. 2006.

PT

Configuration of the Indian Moho beneath the NW Himalaya and Ladakh, J. Geophys. Res., 33, L15308, doi: 10.1029/2006GL026076.

RI

Schärer, U.,Xu, R. H., Allegre, C. J. 1984. U-Pb geochronology of Gangdese, Trans

NU SC

Himalaya, plutonism in the Lhasa-Xigazeregion, Tibet. Earth Planet. Sci.Lett. 69, 311–320.

Schulte-Pelkum, V. et al. 2005. Imaging the Indian subcontinent beneath the Himalaya.

MA

Nature 435, 1222-1225.

Searle, M. P., Pickering, K. T., Cooper, D. J. W., 1990. Restoration and evolution of the intermontane Indus molasses basin, Ladakh Himalaya, India. Tectonophysics 174,

ED

301-314.

CE PT

Searle, M. P., Weinberg, R. F., Dunlap, W. J., 1998. Transpressional tectonics along the Karakoram fault zone: northern Ladakh: Constraints on Tibetan extension. In: Holdsworth,

R.E.,

Sirachan,

R.A.,

and

Dewey,

J.F.,

eds.,

Continental

AC

Transpressional and Transtensional Tectonics. J. Geol. Soc. Lon.135, 307- 325. Sen, K., Das, S., Mukherjee, B.K., Sen, K., 2013. Bimodal stable isotope signatures of Zildat Ophiolitic Mélange, Indus Suture Zone, Himalaya: implications for emplacement of an ophiolitic mélange in a convergent setup, Int. J. Earth Sci. 102, 2033–2042. Sen, K., Mukherjee, B. K., Collins, A. S., 2014. Interplay of deformation and magmatism in the Pangong Transpression Zone, Eastern Ladakh, India: Implications for remobilization of the trans-Himalayan magmatic arc and initiation of the Karakoram Fault. J. Struct. Geol.62. 13-24.

23

ACCEPTED MANUSCRIPT

Sibson, R.H., 1992. Implication of fault-valve behaviour for rupture nucleation and

PT

recurrence. Tectonophysics 211, 283–293. Sinclair, H. D., Jaffey, N., 2001. Sedimentology of the Indus Group, Ladakh, northern India:

RI

implications for the timing of initiation of the palaeo-Indus River. J. Geol. Soc. Lon.

NU SC

158, 151-162.

Sokos, E., Zahradník J. 2008. ISOLA A Fortran code and a Matlab GUI to perform multiplepoint source inversion of seismic data, Comput. Geosci. 34, 967–977. Tapponnier, P., Peltzer, G., Armijo, R., 1996. On the mechanics of the collision between

MA

India and Asia. in: Collision Tectonics, edited by M. P. Coward, and A. C. Ries, J. Geol. Soc. Lon. Sp. 19,115–157, 1996.

ED

Unsworth, M.J.et al. 2005. Crustalrheology of the Himalaya and Southern Tibet inferred from

CE PT

magnetotelluric data.Nature 438, 78–81.http://dx.doi.org/10.1038/nature04154Geol. Soc. Lon. Sp. 19,115–157, 1996. Upadhyay, R., Sinha, A.K., Chandra, R., Rai, H., 1999. Tectonic and magmatic evolution of

AC

the eastern Karakoram, India. Geodyn. Acta 12, 341–358. Valli, F. et al. 2008. New U-Th/Pbconstraintson timing of shearing and long-term slip-rate on the

Karakorum

fault.

Tectonics

27,

TC5007.

http://dx.doi.org/10.1029/2007TC002184. Van Buer, N. J., Jagoutz, O., Upadhyay, R., Guillong, M., 2015. Mid-crustal detachment beneath western Tibet exhumed where conjugate Karakoram and Longmu-Gozha Co faults intersect. Earth Planet. Sci. Lett. 413, 144-157. Waldhauser, F. and Ellsworth, W.L. 2000. A double-difference earthquake location algorithm: Method and application to the Hayward fault, Bull. Seismol. Soc. Am. 90, 1353-1368. 24

ACCEPTED MANUSCRIPT

Waldhauser, F., 2001. hypoDD: A computer program to compute double-difference

PT

hypocenterlocations, U.S. Geol. Surv. Open File Rep. 01-113, 25 pp. Weinberg, R. F., Dunlap, W. J., Whitehouse, M., 2000. New field, structural and

RI

geochronological data from the Shyok and Nubra valleys, northern Ladakh: Linking

NU SC

Kohistan to Tibet. in Tectonics of the Nanga Parbat Syntaxis and the Western Himalaya, edited by T. P. J. Khan et al., J. Geol. Soc. Lon. Sp. 170, 253–275. Weinberg, R. F., Dunlap, W.J.,2000. Growth and deformation of theLadakh Batholith,

MA

northwest Himalayas: implications for timingof continental collision and origin of calc-alkaline batholith.J. Geol. 108, 303–320. Weinberg, R.F., Mark, G., 2008. Magma migration, folding and disaggregation ofmigmatites

ED

in the Karakoram Shear Zone, Ladakh, NW India. Geol. Soc. Am. Bull.120, 994-

CE PT

1009.

Weinberg, R.F., Mark, G., Reichardt, H., 2009. Magma ponding in the Karakoramshear zone, Ladakh, NW India. Geol. Soc. Am. Bull. 121, 278-285. Wittlinger, G.et al. 2004. Teleseismic imaging of subducting lithosphere and Moho offsets

AC

beneath western Tibet. Earth Planet. Sci. Lett. 221, 117-130. Zahradnik, J., Serpetsidaki, A., Sokos, E., and Tselentis, G.A. 2005. Iterative deconvolution of regional waveforms and double-event interpretation of the 2003 Lefkada earthquake, Greece, Bull. Seismol. Soc. Am. 95, 159–172. Zahradník, J., Janský J., and Plicka V. 2008. Detailed waveform inversion for moment tensors of M ∼4 events: Examples from the Corinth Gulf, Greece. Bull. Seismol. Soc. Am. 98(6), 2756–2771, doi 10.1785/0120080124.

25

ACCEPTED MANUSCRIPT

Zhao, J. et al. 2010. The boundary between the Indian and Asian tectonic plates below Tibet.

PT

Proc. Natl. Acad. Sci. 107, 11229-11233. Zhao, W. et al. 2011. Tibetan plate overriding the Asian plate in central and northern Tibet.

AC

CE PT

ED

MA

NU SC

RI

Nat.Geosci. 4, 870-873.

26

ACCEPTED MANUSCRIPT

Figure captions:

PT

Figure 1. (a) Simplified Geological map of northwest part of the Himalaya extending from Tethayan Himalaya, Indus suture zone (ISZ) and Karakoram zone (modified after

RI

Mahéo et al. 2004; Epart and Steck, 2008; Hazarika et al., 2014). The Zildat fault is

NU SC

marked as ZF. The red rectangle shows the study area. The blue and yellow triangles represents seismological stations of Ladakh and Kinnaur network. The distribution of earthquakes are shown by filled circles. The study region is shown by a rectangular box in the inset map of India and adjoining regions at the bottom left

MA

corner. The depth distribution of the earthquakes projected in AB section orthogonal to the NW-SE tectonic features, are shown in (b) where significant tectonic units like

ED

Tso-Morari Crystalline Complex (TMC), Ladakh Batholith (LB) and KarakoramFault-Zone (KFZ) is marked.

CE PT

Figure 2. Example of waveforms fit for a moment tensor solution of an earthquake of Magnitude 4.2 (event date: 04/10/2010, Origin Time: 03h:40m:27.5s; Event No. 5 of Table 1); red waveform is the synthetic, black is the observed, while the waveforms

AC

in gray depict poor fit consequently they are excluded from the calculations in order to determine a final solution. Figure 3. (a) Space position vs time shift correlation plots for the event shown in Fig. 2 (Event No. 5 of Table 1), with maximum correlation normalized to unity. The acceptable solutions are those with largest normalized correlation value. Vertical axis refers to the trial source depth. Horizontal axis refers to the temporal grid search between −3 and +3 seconds with respect to origin time. The largest red beachball is the best-fit solution.

27

ACCEPTED MANUSCRIPT

Figure 4. Variation of correlation values with respect to various (a) trial depths and (b) time

PT

shift are shown. Maximum correlation is observed at 12 km depth with a time shift of 0.35. DC % is observed to be 95.8.

RI

Figure 5. The double-couple percentage (DC%) versus correlation plot is shown which are

NU SC

obtained from grid search during inversion of the event shown in Fig. 2. Cross symbols correspond to moment-tensor calculated for different spatio-temporal grid search. Red symbol refer to the solutions at all trial spatial positions with optimum time shift. The blue symbol shows the double-couple percentage variation with trial

MA

time shift at the optimum spatial position.

Figure 6. 3D block diagram (Geological cross-section along the A-B transect of Fig. 1, not

ED

to scale) showing different litho-tectonic units of the trans-Himalaya and major features of the present day Indian and Tibetan lithosphere. The subsurface features

CE PT

are inferred based on previous studies (Klemperer et al., 2013; Hazarika et al., 2014 and references therein). Transpressive movement along the KF can be seen till lower crustal depth. Shallow depth Indus Thrust demarcates the contact between the Indus

AC

Molasse and the Ladakh Batholith. The Zildat detachment fault marks the contact between the TMC and the ophiolites and the ophiolitic mélange. The shallow subsurface low resistivity zone, caused by presence of fluids, is inferred from magnetotelluric studies of Harinarayana et al., 2006 and Aziz and Harinarayana, 2007. Karzok fault, now activated as a thrust binds the southwestern border of the TMC. It can also be envisaged from this diagram that the deep earthquakes occurring at 30-50 km depth in the TMC region is occurring near the interface of mid-crustal low velocity zone and relatively more competent Indian lower crust.

28

ACCEPTED MANUSCRIPT

Table 1: List of hypocentral parameters of the earthquakes used for waveform inversion

PT

along with estimated Moment Magnitude and inferred fault plane solutions. The number of stations used for inversion, average double couple percentage and

AC

CE PT

ED

MA

NU SC

RI

variance reduction during waveform inversion are mentioned for each earthquake.

29

AC

CE PT

ED

MA

NU SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 1 30

CE PT AC

Figure 2

ED

MA

NU SC

RI

PT

ACCEPTED MANUSCRIPT

31

AC

Figure 3

CE PT

ED

MA

NU SC

RI

PT

ACCEPTED MANUSCRIPT

32

AC

CE PT

ED

MA

NU SC

RI

PT

ACCEPTED MANUSCRIPT

Figure 4 33

AC

Figure 5

CE PT

ED

MA

NU SC

RI

PT

ACCEPTED MANUSCRIPT

34

CE PT AC

Figure 6

ED

MA

NU SC

RI

PT

ACCEPTED MANUSCRIPT

35

ACCEPTED MANUSCRIPT

List of hypocentral parameters of the earthquakes used for waveform inversion

PT

Table 1:

RI

along with estimated moment magnitude and inferred fault plane solutions. The

NU SC

number of stations used for inversion, average double couple percentage and variance reduction during waveform inversion are mentioned for each earthquake. SL.

Date

No.

(yy/mm/dd)

Origin Time

Latitude Longitude Depth (˚N)

(˚E)

(Km)

Parameters

(Mw)

of

2009/10/19 15:04:53.1

33.523

2

2010/01/03 20:05:13.5 32.906

3

2010/02/20 20:52:50.2 32.961

4

inferred fault plane

No. of Average Variance stations Double Reduction used couple (%) percentage (%)

Strike Dip Rake (˚) (˚) (˚)

78.752

12.7

4.1

299

71

152

3

93.0

73

78.272

9.9

3.5

255

52

79

6

79.5

49

78.375

9.6

3.8

105

53

51

7

78.0

57

2010/08/14 03:08:17.6 34.124

78.735

66.0

4.1

301

58

157

3

75.8

59

5

2010/10/04 03:40:27.5 33.277

78.240

12.3

4.2

341

26

-94

8

95.8

63

6

2011/04/21 19:31:14.2 33.879

78.417

10.0

4.1

339

80

123

6

89.5

69

7

2011/07/10 00:26:33.1 33.854

78.335

14.1

4.3

123

80

142

3

68.0

51

8

2011/08/04 01:23:11.0 33.340

78.331

13.4

3.6

270

54

59

2

72.4

57

9

2011/12/03 09:59:00.2 33.263

78.263

13.7

3.6

255

57

111

4

76.7

78

10 2012/01/16 16:50:43.5 33.857

77.833

7.8

3.0

135

50

89

3

70.0

65

11 2012/02/17 02:29:41.3 34.232

78.294

21.3

3.7

158

85

126

3

69.0

59

12 2012/04/24 01:52:45.0 34.162

78.042

13.1

3.9

300

86

148

3

78.0

58

13 2012/04/25 01:40:02.4 34.165

78.016

13.2

3.5

150

77

152

3

79.0

51

AC

CE PT

1

ED

MA

(hh:mm:ss)

Magnitude

36

ACCEPTED MANUSCRIPT

Highlights

PT

Seismotectonics of trans-Himalaya is inferred through local seismicity

RI

The Karakoram Fault is transpressional and penetrates at least up to the lower crust

NU SC

The exhumation of Tso-Morari dome is by combination of detachment and thrusting

AC

CE PT

ED

MA

Fluid may play major role in seismicity in Zildat Detachment zone

37