- Email: [email protected]
Earthquake genesis in Nepal Himalaya: A perspective from imaging of the 25th April 2015 Mw 7.8 earthquake source zone
Accepted Manuscript Earthquake genesis in Nepal Himalaya: A perspective from imaging of the 25th April 2015 Mw 7.8 earthquake source zone Anand K. Pandey, Dipankar Saikia, M. Ravi Kumar PII: DOI: Reference:
S1367-9120(16)30450-3 http://dx.doi.org/10.1016/j.jseaes.2016.12.039 JAES 2908
To appear in:
Journal of Asian Earth Sciences
Received Date: Revised Date: Accepted Date:
23 November 2015 28 December 2016 28 December 2016
Please cite this article as: Pandey, A.K., Saikia, D., Ravi Kumar, M., Earthquake genesis in Nepal Himalaya: A perspective from imaging of the 25th April 2015 Mw 7.8 earthquake source zone, Journal of Asian Earth Sciences (2016), doi: http://dx.doi.org/10.1016/j.jseaes.2016.12.039
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.
Earthquake genesis in Nepal Himalaya: A perspective from imaging of the 25th April 2015 Mw 7.8 earthquake source zone Anand K. Pandeya, Dipankar Saikia b, M. Ravi Kumara,c* a CSIR-National Geophysical Research Institute, Hyderabad, India b Indian National Center for Oce4an Information Services, Hyderabad, India c Institute of Seismological Research, Gandhinagar, India ---* Corresponding Author: M. Ravi Kumar Email address: [email protected]
Abstract The Mw 7.8 earthquake in Central Nepal nucleated in the mid-crustal ramp zone of the Main Himalayan décollement Thrust (MHT) and propagated eastward for >140 km where the largest triggered event of Mw 7.3 occurred without any surface rupture. Although it is advocated that the slip and rupture dynamics are controlled by the structural configuration of the MHT and the upper crust, precise correlation between the seismic structure and seismogenesis is hitherto scarce in the Himalaya. To address the issue, we imaged the crustal structure along three profiles covering the earthquake source region using receiver function analysis of the seismic data from the HiCLIMB and HIMNT seismic networks to understand the lateral variability. A ~5 km thick, low velocity layer is observed at the mid–crustal level, that steepens in the MHT ramp zone. The bulk of the seismicity including large shocks after the 2015 Nepal earthquake lies in the vicinity of this low velocity layer. Correlation of the seismic structure and aftershock distribution with the published crustal structure clearly suggests that the rupture involves a thicker zone extending for >40 km to the south of the source zone in the MHT ramp. We refined the structure of the MHT zone incorporating published coseismic slip and ground deformation to suggest that the rupture terminated at the footwall imbricate (horse) on the floor thrust below the zone of maximum coseismic uplift and there was a two stage rupture towards the eastern margin of the rupture zone. Key words: Mw 7.8 Nepal Earthquake, Receiver functions, Crustal structure, Mid-crustal ramp 1. Introduction The 25 April 2015 earthquake of magnitude (Mw) 7.8 that occurred on a gently (710°) north dipping, near E–W trending thrust fault, at about 15 km depth (Fig. 1; USGS, 2015; Galetzka et al., 2015), is the largest event recorded in the Nepal Himalaya during the modern instrumental era. The rupture propagated eastward within ~40 seconds (Avouac et
al., 2015; Yagi and Okuwaki, 2015) where most of the aftershocks were clustered and the largest aftershock/triggered earthquake of Mw 7.3 occurred on 12th May, 2015 at a depth of 17 km. Over 500 local earthquakes / aftershocks with magnitude >4.0 occurred within 45 days of the main event (Fig. 1; USGS, 2015; Adhikari et al., 2015). It is construed that the hypocentre lies within the mid crustal ramp on the Main Himalayan Thrust (MHT) (Avouac et al., 2015), which is well established in this region (Ni and Barazangi, 1984; Schelling, 1992; Pearson and DeCelles, 2005; Avouac, 2015). The spatial distribution of the aftershocks defines a nearly 140 x 40 km2 rectangular fault zone that ruptured during the 2015 Nepal earthquakes (Fig. 1) abutting the presumed extension of the rupture zone of the 1934-M 8.4 Great Bihar Nepal earthquake (Adhikari et al., 2015). Results from InSAR line-of-sight displacement data gleaned from the ALOS-2 data show that the slip during the main shock and the largest aftershock/triggered event extends for over >140 km with a peak slip of 5.5 – 6.5 m on the MHT between 5 and 15 km depth and a >1m co-seismic uplift just north of Kathmandu and a corresponding depression on either side of the uplift zone (Lindsey et al., 2015; Avouac et al., 2015; Elliott et al., 2016; Sreejith et al., 2016). The main shock and the aftershocks of the 2015-Nepal earthquake did not produce any surface rupture and the observed ground deformation (Lindsey et al., 2015; Sreejith et al., 2016; Elliot et al., 2016) largely remains confined to the Lesser Himalayan region, which constitutes the locked zone of the MHT having an interseismic strain deficit rate of around 17.8 ± 1.5 mm/yr (Ader et al., 2012; Bilham et a., 1997) for over a decade in the region. The seismogenesis in the Himalayan region is controlled by the mid–crustal ramp on the MHT, which is the source zone of the earthquakes and microseismicity (Pandey et al., 1999; Schulte–Pelkum et al., 2005; Avouac, 2015), though its configuration is not well constrained. The large and great earthquakes rupture the southern flat of the MHT and emerge towards the Himalayan front depending on the co–seismic slip distribution and the geometry of the structures along the
rupture zone (Mugnier et al., 2013; Bollinger et al., 2014; Avouac, 2015). The 25th April, 2015 earthquake affected zone also experienced the Mw 7.6 – 1833 Nepal earthquake (Bilham, 1995; Mugnier et al., 2013) and 1934-Bihar Nepal Great earthquake, which produced surface rupture and ground deformation along the MFT in the adjoining region (Sapkota et al., 2013; Bollinger et al., 2014). In this paper, we performed common conversion point (CCP) receiver function analysis of the HiCLIMB (Nábělek et al., 2009) and HIMNT (Schulte-Pelkum et al., 2005) data along three sections AB- traversing the epicentre of the Mw 7.8 main event, CD- across Kathmandu, where maximum ground deformation is observed, and EF – across the epicentre of the largest aftershock/ triggered earthquake (Fig. 1). We correlated seismic receiver function analysis results with the crustal structures along the respective sections (Pearson & DeCelles, 2005; Schelling, 1992) and integrated them with the coseismic geodetic observations (Lindsey et al., 2015) to understand the seismogenesis and rupture during the 2015 Nepal earthquake. 2. Receiver function imaging of the crust The data accrued from the HiCLIMB and HIMNT networks offered an excellent opportunity to interrogate and validate the structure of the seismogenic zones in the Himalayan mountain belt using the receiver function (RF) analysis. Waveforms used in this study are extracted from the broadband stations deployed under the HiCLIMB and HIMNT seismological experiments in the Nepal Himalaya and southern Tibet and archived in the IRIS Data Management Centre. However, only a subset of data from stations of the HiCLIMB experiment, located within our study area south of 30ºN latitude is utilized (Fig. 1). Three component seismograms due to earthquakes with magnitude >5.5 in the epicentral distance range 30º to 90º having a P-wave SNR > 3 are selected for computation of receiver functions. A total of 1535 and 533 events that yielded good quality waveforms have a
reasonably good azimuthal coverage at the HiCLIMB and HIMNT stations, respectively (Fig. 2). Receiver function analysis is a powerful tool to investigate the receive side subsurface structure. The receiver function method is based on the fact that the coda of the Pwave contains the P to S (Ps) conversions and their reverberations resulting from the impedance contrast across layers beneath the receiver. These Ps conversions and reverberations can be isolated through deconvolution of the vertical component from the radial component of the seismic record of a teleseismic earthquake. However, the simple concept of deconvolution is tricky to implement due to instability arising from division by zero. To avoid this instability, several methods of receiver function calculation, like the spectral water level deconvolution (Langston, 1979; Owens et al., 1983; Ammon et al., 1990), deconvolution in the time domain by least-squares estimation (Abers et al., 1995), iterative time domain deconvolution (Ligorría and Ammon, 1999) and multitaper frequencydomain cross-correlation receiver function (MTRF) (Park and Levin, 2000), have been developed. The advantage of MTRF method over the other methods is the use of multitapers to minimize spectral leakage during estimation of the spectrum and use of a pre-event noise spectrum for frequency dependent damping. A drawback of this method is that it uses a short analysis window, which limits its usability in situations where information at longer lags is desired. In this study, the receiver functions are calculated using the extended-time multitaper frequency domain technique (EMTRF) of Helffrich (2006), which is an extension of the MRTF method of Park and Levin (2000) for computing receiver functions of arbitrary length. This method uses a series of short, overlapping, multiple tapers to window the time series across its length and sums the individual Fourier transformed signals to produce a receiver function estimate which preserves the phase information for each sub-window. The vertical and horizontal components of the seismograms are decomposed into P, Sv and Sh
components using the back azimuth and incidence angle estimated from the first 10 s of the seismograms around the P–wave arrival prior to applying the ETMTRF. A frequency domain low–pass cos2 taper with a cut off frequency at 1Hz is applied to the all the RFs to avoid Gibbs effect. After an initial quality check, 25,000 high quality RFs from 1848 events are retained for further processing. The common conversion point (CCP) stacking method has been used to image the structure along linear profiles shown in Fig. 1. The receiver functions are migrated to depth using Ps travel time equation and the ak135 velocity model whose crustal part is replaced by crust 1.0 (Laske et al., 2013) velocity model based on the location of the station. The depth migrated receiver function amplitudes are then projected onto a vertical plane along profiles AB, CD, EF covering the entire rupture zone of the 2015-Nepal earthquake (Fig. 1) and binned in 1x1 km grids using Fresnel zone approximation, summed and normalized by the total no RF’s in each grid (Fig. 3). The CCP image produced by this method along the profile CD (Fig. 3b) is slightly different from the that produced by Nábělek et al., (2009) in the same region. The difference is due to the fact that Nábělek et al., (2009) used a method of stacking migrated images produced using travel time equation for direct conversions (i.e Ps) and the multiples (i.e PpPs and PpSs+PsPs). Although, the stacking of migrated images for Ps and its multiples greatly enhances the Moho signal, the shallow structure may contain artefacts due to migration of Ps energy to shallow depths while migrating for the multiples (Wilson et al., 2005). Since the emphasis of this study is not on mapping the Moho discontinuity but on mapping the other shallow interfaces within the crust, our approach is more appropriate. Duputel et al., (2016) have also imaged the MHT zone along Hi-CLIMB network (C-D profile of present study) using P and S receiver functions computed using the iterative time domain deconvolution method (Ligorría and Ammon, 1999) and CCP stacking using a 1D velocity model of Grandin et al. (2015). Also, using a 1D velocity model in a region where large lateral variations in velocity structure are expected,
the CCP image produced by them may not necessarily represent the true shallow sub-surface picture in the main shock region. We try to overcome this issue by replacing the crust beneath each station with the relevant model from crust 1.0 and perform the CCP imaging. 3. Seismic structure and seismogenesis The CCP stacks are overlaid with the aftershock distribution and fault plain solutions of >M 5 events derived from USGS catalogue (Fig. 3). To understand the seismogenesis in Nepal, we interpreted the seismic images in conjunction with focal mechanisms of the Mw > 5 earthquakes that occurred during April–May, 2015 and the published crustal structures along sections CD (after Pearson & DeCelles, 2005) and EF (Schelling, 1992; Figs. 4). Since the seismic station coverage along section A-B passing through the main shock is sparse (Fig. 1), the crustal structure could not be imaged with a good resolution (Fig. 3a). The Mw 7.8 event occurred at ~12 km depth in a ~7-10° dipping MHT ramp zone (USGS; Galetzka et al., 2015; Adhikari et al., 2015) that is associated with a tabular low velocity layer, which has a diffuse but continuous southward extension along with the bulk of aftershocks in its vicinity (Fig. 3a). Based on map pattern and aftershock distribution, the generalized cross section with the extent of rupture is depicted (Fig. 4a). The dense network of HiCLIMB profile (Fig. 1) yielded a well constrained seismic section of the earthquake affected region passing through Kathmandu C-D (Fig. 3b). The CCP image reveals a crustal structure with higher velocities in the upper 7-10 km, underlain by a >5 km thick low velocity layer down to 10–15 km depth in the upper crust. Interestingly, most of the large aftershocks (Mw > 5) are confined to the low velocity layer in the Lesser Himalayan zone. This upper crustal low velocity layer was observed in the HiCLIMB section by Nábělek et al. (2009) but was not specifically discussed. The shallow high velocity structure can be correlated with the highly deformed Lesser Himalayan Crystallines and the underlying low velocity layer with the Lesser Himalayan sedimentary sequence as well as the highly sheared MHT zone (Fig. 4). This flat low velocity
slab deepens beneath the rising Himalayan topography to the south of the Main Central Thrust (MCT) zone, demarcating the ramp on the MHT where the large earthquake events have occurred (Fig. 3a, 5a) and the background seismicity is clustered (Pandey et al., 1999). Similar inferences were drawn in the Garhwal region where a narrow low velocity layer demarcates the MHT with a well defined ramp that coincides with the local seismicity observed during the 1999- Chamoli earthquake (Caldwell et al., 2013). To understand the structural control on seismogenesis, we overlay retro-deformable balance cross section by Pearson & DeCelles, (2005) over the seismic image (Fig. 4b). Unlike the Garhwal Himalaya, the tabular low velocity layer with internal perturbations along the CD section in Nepal does not correspond well to the MHT marked by Pearson and DeCelles, (2005; Fig. 4b). The low velocity zone is wider and a large number of aftershocks occur below the MHT flat (Fig. 4b). Further, the location of the ramp on the décollement and the Lesser Himalayan duplex in the ramp zone along section CD remain ambiguous and show a vague correlation with contrasting velocity layers (Fig. 4b). After the Nepal earthquake, Duputel et al. (2016) have also reprocessed the HiCLIMB data along C-D section identifying a simple MHT flat and the ramp at the base of low velocity layer where aftershocks are clustered. It was observed during the 2015 Nepal earthquake that the rupture propagated eastward and decelerated after 40 seconds (Avouac et al., 2015; Yagi and Okuwaki, 2015; Adhikari et al., 2015). The bulk of the aftershocks, including the largest aftershock/triggered earthquake of Mw 7.3 on 12th May 2015, are confined to the deceleration zone towards the eastern side of the rupture zone (Fig. 1) and the larger aftershocks are situated at deeper depths compared to those in the western margin. The seismic and crustal structures along the AB and CD sections (Figs. 4a, 4b) show a lateral variation. Since the mid-crustal ramp on the MHT is considered the earthquake source zone, the lateral variations in its geometry have profound influence on the seismogenesis. To understand the lateral variation in crustal
structure, we also generated a common conversion point image of receiver functions along profile E-F that samples the eastern side of the rupture zone (Fig. 1), using data from stations of HIMNT seismic network and overlaid the crustal structure and litho-tectonic disposition after Schelling (1992; Fig. 4c). Albeit having a low resolution owing to the sparse density of stations, the seismic section reveals crustal structure (Fig. 3c) that is distinctly different from that below the western segment (Figs. 3a, b) of the rupture zone. The projected aftershocks including the M 7.3 event are clustered in a relatively narrow zone in the vicinity of the low velocity layer but are located below the MHT and to the south of the ramp zone modelled by Schelling (1992; Fig. 4c). The earlier experiment along the HIMNT seismic section (~70 km east of the EF section) mapped the MHT as a diffused zone of seismic anisotropy and could not clearly image the mid crustal ramp on the MHT (Schulte-Pelkum et al., 2005). These observations warrant inclusion of a seismogenic low velocity layer and the zone of aftershocks in the Himalayan wedge to explain seismogenesis, in spite of the limitation due to sparse station coverage and poor seismic structure. 4. Coseismic slip and uplift pattern vis-a-vis lateral variability in the rupture The Mw 7.8 earthquake ruptured the mid-crustal MHT for more than 140 km, with the largest aftershock/triggered event of Mw 7.3 defining the eastern limit (USGS; Fig. 1). Results of InSAR interferometry from the earthquake affected region show coseismic ground deformation for over 100 km along a strike parallel narrow zone to the north of Kathmandu with maximum >1 m uplift and a corresponding depression towards north during the Mw 7.8 event (Fig. 5a, 5b) and ~50 cm uplift associated with the Mw 7.3 event (Fig. 5c; Lindsey et al., 2015; Avouac et al., 2015; Elliot et al., 2016; Sreejith et al., 2016). The modelled coseismic slip on the décollement suggests that the maximum slip of >5m on the mid-crustal MHT within the Lesser Himalaya and the maximum coseismic uplift lie to the south of the zone of maximum coseismic slip (Fig. 5a). The coseismic slip and uplift are clearly bi model
towards the eastern side of the rupture zone, where the Mw 7.3 event possibly released the residual strain with limited spatial extent (Fig. 1, 5). Interestingly, the aftershock pattern during the Nepal earthquake shows two zones of seismicity i.e. along the main event possibly in the source zone and another towards south beneath the Kathmandu basin (Adhikari et al., 2015). We compared the observed spatial extent of the coseismic uplift and the modelled coseismic slip on the décollement (Fig. 5; Lindsey et al., 2015) with the interpreted crustal structures along CD and EF sections (Fig. 6) and aftershock distribution pattern (Adhikar et al., 2015), to understand the rupture propagation. The multiple imbricates and the frontal horse emanating from a deeper (20-25 km depth) décollement (Pearson and DeCelles, 2005) do not conform with the modelled seismic structure and the aftershock distribution of the 2015 earthquake (Fig. 4b). We refined the mid-crustal structure incorporating the new observations (Fig. 6b). The steepening of low velocity slabs in the seismic section (Fig. 6b) beneath the Lesser Himalayan duplex, clearly define the ramp zone that flattens to the basal décollement at >15 km depth, possibly marking the southern extent of the aseismic segment of the MHT. The redefined MHT ramp connects to the southern flat, which encompasses the aftershock cluster and the velocity interface without disturbing the shallower structural configuration (Fig. 6b). This basal décollement together with the previously defined MHT flat (Pearson & DeCelles, 2005) forms a duplex zone encompassing most of the seismicity during the 2015 event. The aftershocks terminate at the zone of maximum co-seismic uplift, which possibly defines the frontal ramp/horse of the duplex structure on the basal décollement that ruptured during Mw 7.8 event (Fig. 6b). However, towards the eastern margin of the rupture zone, the MHT configuration (Schelling, 1992) does not confirm with the seismogenesis as the bulk of seismicity lies below the décollement (Fig. 4c). This zone experienced a two stage coseismic uplift of ~0.5 m each during the Mw 7.8 and Mw 7.3 events (Figs. 5a, 5c; Lindsey et al., 2015) possibly suggesting
multiple rupture during the respective events. The Mw 7.8 event produced ~0.5 m uplift corresponding to ~2- 3 m of modelled coseismic slip (Fig. 5a, 5c); whereas the Mw 7.3 event produced ~0.5 m uplift along the MHT ramp zone with >5 m peak coseismic slip (Fig. 5a; Lindsey et al., 2015) during the triggered rupture (Fig. 6c) accommodating slip deficit towards the eastern margin. We refined the MHT configuration primarily encompassing the aftershock distribution trend and the respective extent of ruptures during the Mw 7.8 and 7.3 events (Fig. 6c). This refined MHT configuration of the rupture zone along with that of Schelling (1992) forms a duplex in MHT zone with multiples imbricates / horses. Similar horse structure in the MHT (duplex) zone is proposed to explain the wider microseismic distribution and topography proxies in western Nepal (Harvey et al., 2015). The other unique structure along the EF section is the out-of-sequence Sun Kosi Thrust (SKT; Fig. 5b), which lies to the immediate south of the zone of maximum coseismic uplift (Fig. 1) and aftershock occurrence, without any noticeable co-seismic deformation during the 2015 Nepal earthquake. This suggests that the earthquake rupture terminates before the SKT imbricate on the MHT and probably within the MHT duplex at a greater depth than in the western flank (Fig. 6). We tried to analyse the coseismic slip and uplift distribution during the 2015 earthquake (Fig. 5) together with the interpreted MHT structure (Fig. 6). To accommodate bulk seismic slip during a thrust earthquake event without appreciable terrain uplift, it is imperative that the decollement thrust be wider, gently dipping, producing no significant throw (vertical) component during faulting (Fig. 6b, 6c) and/or the strain is accommodated by internal deformation within the duplex. The same is evident from the wider aftershocks distribution to the south of the MHT ramp (Fig. 6). However, the localized terrain uplift away from the peak coseismic slip (Fig. 5a, 5b) points towards steeply dipping thrust producing a larger throw (vertical component) and focussed uplift. Since, no surface rupture was observed
after the earthquake and no corresponding structure is exposed on the ground along the uplift zone, it possibly points to a blind thrust as an imbricate / horse on the MHT flat underneath the zone of maximum surface uplift causing linear surface uplift during rupture (Figs. 5, 6). Gehrels et al. (2003) have indicated front parallel thrusts within the Kathmandu nappe (Fig. 4b) and the out-of-sequence Sun Kosi Thrust (SKT; Figs. 1, 4c; Schelling (1992), to the east of the rupture zone which remains blind beneath the Kathmandu nappe; but no coseismic deformation is mapped or reported along these structures and their relation to the seismogenesis during the 2015 Nepal earthquake is not clear due to insufficient observations. 5. Discussion The Himalaya of Central Nepal has experienced several large earthquakes during the last millennium. Notable among them are the earthquakes in 1934, 1833, ~1255, and ~1100 AD, that severely affected the Kathmandu valley despite being at varied distance from the epicentres (Pandey and Molnar, 1988; Bilham, 1995; Mugnier et al., 2013; Sapkota et al., 2013; Bollinger et al., 2014). The rupture zone of the 2015 earthquake (Fig. 1; Avouac et al., 2015; Yagi and Okuwaki, 2015) overlies the relocated rupture zone of the Mw ~7.6 – 1833 earthquake (Bilham, 1995; Ambrasey and Douglas, 2004; Mugnier et al., 2013). The proximity of the relocated epicentre of the 1833 earthquake (Mugnier et al., 2013) to the largest aftershock (Mw 7.3) of the 2015 event suggests an interseismic interval of 182 years for the rupture on the same segment of the MHT with >5m of coseismic slip without any surface rupture. This suggests ~3.6 m of interseismic strain accumulation considering a ~20 mm/yr slip deficit on the locked segment of MHT to the south of the ramp (Bilham et al., 1997; Ader et al., 2012). The interseismic strain accumulation since 1833 along with the background strain and tectonic loading due to the great earthquake of 1934, which lies to the east of the present rupture zone, match closely with the relieved strain during the 2015-Nepal earthquake.
It is important to address the above ambiguity in the earthquake source zone and rupture in light of the seismic structure (Fig. 6) as the interseismic stress build-up triggers large earthquakes in the ramp zone and rupture the southern flat of the MHT. A comparison of the seismic structures, aftershock distribution pattern, published and interpreted crustal structures along the three profiles (Figs. 3,4,5,6) reveals lateral variation in the seismogenic structure of the 2015-Nepal earthquake zone, from west to east. The invariable association of the low velocity layer, albeit having a poor resolution along the EF section, suggests duplex structures in the MHT zone that accommodate strain during earthquake rupture propagation. However, this inference is under the assumption that the aftershocks are reasonably well located. This brings forth the focus on structural variations along the strike (lateral) in different segments of the Himalaya to explain varying recurrence intervals and rupture behaviour (Mugnier et al., 2013). Using a wider zone of seismicity and topography as proxies, it is argued that the region with a large flat or a wider duplex on the MHT in Lesser Himalayan zone experienced larger interseismic interval (Harvey et al., 2015). In contrast, the crustal structure in eastern Nepal is less complex with a large antiform growth over a sharp ramp on the MHT, which coincides with the zone of microseismicity and earthquake nucleation at mid crustal (~15-20 km) depths (Avouac, 2015). The hypocenters of the Mw 7.8 and 7.3 earthquakes were indeed located in this zone but the aftershocks have a wider distribution along the low velocity layer towards south of the ramp zone (Fig. 3). The low velocity is possibly caused by the lower density composition or higher fluid concentration in the rocks that retard the seismic wave velocity. The low velocity slab possibly represents the Lesser Himalayan sedimentary sequence primarily composed of limestone, sandstone, shale and quartzite (Schelling, 1992; Upreti, 1999). Further, the increasing complexity of mid-crustal structure, i.e. a simpler MHT ramp zone along AB section to the thickened duplex zone on the MHT towards the east (Fig. 4, 6c,
5), enhances the strength of the earthquake source zone that may be responsible for the increasing depth of large events from ~12 (Mw 7.8 event) to ~18 km (Mw 7.3 event) and the retardation of eastward rupture propagation after ~40-45s (Avouac et al., 2015; Yagi and Okuwaki, 2015). The lateral variations in structure may cause heterogeneity in the fault frictional behaviour affecting diminishing co-seismic ground deformation and rupture propagation towards east and lack of southward rupture during the Mw 7.3 aftershock (Fig. 6). 6. Conclusions We synthesize the salient observations and interpretations in a schematic diagram (Fig. 7). The decreasing thickness, increasing depth of the low velocity layer, increasing density and depth of aftershocks towards east and the interpreted structure on the seismic profiles (Figs. 3, 6) in the earthquake rupture zone suggest lateral variation in the upper crustal structure and the MHT configuration. The rupture propagated eastward from the locus of the Mw 7.8 earthquake and decelerated after ~40-45 seconds (Avouac et al., 2015; Yagi and Okuwaki, 2015) due to the duplex on the MHT, which strengthens the décollement zone towards east. The coseismic uplift of ~1 m (Lindsey et al., 2015) to the south of the zone of maximum (>5m) slip, along a strike parallel zone to the north of Kathmandu, marks the southern termination of rupture front on the blind thrust, which is possibly a steeper frontal ramp / horse on the MHT in the duplex zone, providing a weak zone for focused uplift during the 2015 earthquake. The same may be responsible for the repeated ground deformation and associated damage to the Kathmandu valley (Bilham, 1995; Pandey and Molnar, 1988; Ambraseys and Douglas, 2004; Mugnier et al., 2013; Sapkota et al., 2013; Bollinger et al., 2014) during past historical earthquakes. The eastern segment has produced composite rupture with the Mw 7.3 events filling the slip deficit in the rupture zone. Acknowledgements:
Authors acknowledge funding from the GENIAS and HEART programs of CSIRNGRI. IRIS-DMC is gratefully acknowledged for making the data available. We highly appreciate critical and constructive comments from guest editor J. P. Mugnier, Roland Burgmann and an anonymous reviewer that significantly improved the manuscript. References Abers, G.A., Hu, X. and Sykes, L.R., 1995. Sourch scaling of earthquakes in the Shumagin region, Alaska: time-domain inversions of regional waveforms. Geophysical J. Int., 123(1), 41-58. Ader, T., Avouac, J.P., Liu-Zeng, J., Lyon-Caen, H., Bollinger, L., Galetzka, J., Genrich, J., Thomas, M., Chanard, K., Sapkota, S., Rajaure, S., Shrestha, P., Ding, L., Flouzat, M., 2012. Convergence rate across the Nepal Himalaya and interseismic coupling on the Main Himalayan Thrust: Implications: Journal of Geophysical Research, 117, B04403, doi:10.1029/2011JB009071. Adhikari, L.B., Gautam, U.P., Koirala, B.P., Bhattarai, M., Kandel, T., Gupta, R.M., Timsina, C., Maharjan, N., Maharjan, K., Dahal, T., Hoste-Colomer, R., Cano, Y., Dandine, M., Guilhem, A., Merrer, S., Roudil, P., Bollinger, L., 2015. The aftershock sequence of the 2015 April 25 Gorkha–Nepal earthquake. Geophys. J. Int., 203, 2119–2124 Ambraseys, N. N., Douglas, J., 2004. Magnitude calibration of north Indian earthquakes: Geophys. J. Int., 159, 165-206. Ammon, C. J., Randall, G. E., Zandt, G., 1990. On the non-uniqueness of receiver function inversions. J. Geophys. Res., 95, 15,123–15,127. Avouac J.-P., 2015. Mountain Building: From Earthquakes to Geologic Deformation: In Gerald Schubert (Ed.) Treatise on Geophysics, 2nd ed., 6, Oxford: Elsevier, pp. 381-432.
Avouac, J.-P., Meng, L., Wei, S., Wang, T., Ampuero, J.-P., 2015. Lower edge of locked Main Himalayan Thrust unzipped by the 2015 Gorkha earthquake: Nature Geoscience, DOI: 10.1038/NGEO2518. Bilham,R., 1995. Location and magnitude of the 1833 Nepal earthquake and its relation to the rupture zones of contiguous great Himalayan earthquakes. Current Science, 69,101-128. Bilham, R., K. Larson, J. Freymueller, Project Idylhim members, 1997. GPS measurements of present day convergence across the Nepal Himalaya. Nature, 386, 61-64. Bollinger, L., S. N. Sapkota, P. Tapponnier, Y. Klinger, M. Rizza, J. Vander Woerd, D. R. Tiwari, R. Pandey, A. Bitri, S. Bes de Berc, 2014. Estimating the return times of great Himalayan earthquakes in eastern Nepal: Evidence from the Patu and Bardibas strands of the Main Frontal Thrust: Jour. Geophy. Res., 119, doi:10.1002/2014JB010970. Caldwell, W.B., Klemperer, S.L., Lawrence, J.F., Rai, S.S., Ashish, 2013. Characterizing the Main Himalayan Thrust in the Garhwal Himalaya, India with receiver function CCP stacking. Earth and Planetary Science Letters, 367, 15–27. Cattin, R., Avouac, J.P., 2000. Modeling mountain building and the seismic cycle in the Himalaya of Nepal: Journal of Geophysical Research, 105, 13389-13407. Duputel, Z., Vergne,J., Rivera L., Wittlinger, G., Farra, V., György H., 2016. The 2015 Gorkha earthquake: A large event illuminating the Main Himalayan Thrust fault. Geophy. Res. Lett., 43, 2517-2525. DOI: 10.1002/2016GL068083 Elliott, J.R., Jolivet, R., González, P.J., Avouac, J.-P., Hollingsworth, J., Searle, M.P., Stevens, V.L., 2016. Himalayan megathrust geometry and relation to topography revealed by the Gorkha earthquake. Nature Geoscience, 9, 174-180. DOI: 10.1038/NGEO2623. Galetzka, J., et al. (31 authors), 2015. Slip pulse and resonance of Kathmandu basin during the 2015 Mw 7.8 Gorkha earthquake, Nepal imaged with geodesy. Science, doi:10.1126/science.aac6383.
Gehrels, G.E., DeCelles, P.G., Martin, A., Ojha, T.P., Pinhassi, G., Upreti, B.N., 2003. Initiation of the Himalayan orogen as an early Palaeozoic thin skinned thrust belt. GSA Today, 13, 4–9. Grandin, R., M. Vallée, C. Satriano, R. Lacassin, Y. Klinger, M. Simoes, and L. Bollinger, 2015. Rupture process of the Mw = 7.9 2015 Gorkha earthquake (Nepal): Insights into Himalayan
megathrust
segmentation,
Geophys.
Res.
Lett.,
42,
8373–8382,
doi:10.1002/2015GL066044. Harvey, J. E., Burbank, D.W., Bookhagen, B. 2015. Along-strike changes in Himalayan thrust geometry: Topographic and tectonic discontinuities in western Nepal. Lithosphere, 7(5), 511–518. Helffrich, G., 2006. Extended-time multi-taper frequency domain cross correlation receiver function estimation. Bull. Seismol. Soc. Am.,. 96, 344-347. Jackson, M., Bilham, R., 1994. Constraints on Himalayan deformation inferred from vertical velocity fields in Nepal and Tibet: Jour. Geophy. Res., 375 99(B7), 13897-13912. Langston, C.A., 1979. Structure under Mount Rainier, Washington, inferred from teleseismic body waves. J. Geophys. Res., 84, 4749–4762. Laske, G., Masters., G., Ma, Z., Pasyanos, M., 2013. Update on CRUST1.0 -A 1-degree Global Model of Earth’s Crust: Geophys. Res. Abstracts 15, Abstract EGU2013-2658. Ligorría, J.P., Ammon, C.J., 1999. Iterat ive deconvolut ion and receiver function estimation. Bull. Seismol. Soc. Am., 89, 1395–1400 Lindsey, E.O., Natsuaki, R., Xu, X., Shimada, M., Hashimoto, M., Melgar, D., Sandwell, D. T., 2015. Line-of-sight displacement from ALOS-2 interferometry: Mw 7.8 Gorkha Earthquake and Mw 7.3 aftershock. Geophy. Res. Lett., 42, 6655-6661.
Mugnier, J. L., Gajurel, A., Huyghe, P., Jayangondaperumal, R., Jouanne, F., Upreti, B. N., 2013. Structural interpretation of the great earthquakes of the last millennium in the central Himalaya: Earth Science Review,127, 30-47. Nábělek, J., Hetnyi, G., Vergne, J., Sapkota, S., Kafle, B., Mei Jiang, Su, H., Chen,J., Huang, B.-S, the Hi-CLIMB Team, 2009. Underplating in the Himalaya-Tibet Collision Zone Revealed by the Hi-CLIMB Experiment. Science, 325, 1371-1374. Ni, J., Barazangi, M., 1984. Seismotectonics of the Himalayan collision zone: geometry of the under thrusting Indian plate beneath the Himalaya: Journal of Geophysical Research, 89, 1147-1163. Owens, T.J., Taylor, S. R., Zandt, G., 1983. Isolation and enhancement of the response of local seismic structure from teleseismic P-wave forms. Lawrence Livermore Lab. Report UCID 19809, 1-33. Pandey, M. R., Molnar P., 1988. The distribution of intensity of the Bihar-Nepal earthquake of 15 January 1934 and bounds on the extent of the rupture zone: J. Nepal Geological Society, 5(1), 22-44. Pandey, M.R., Tankudar, R.P. Avouac, J.-P., Vergne, J., Hritier T., 1999. Seismotectonics of the Nepal Himalaya from a local seismic network: Jour. Asian Earth Sci., 17, 703-712. Park, J., Levin, V., 2000. Receiver functions from multiple-taper spectral correlation estimates. Bull. Seismo. Soc Amer., 90, 1507-1520. Pearson, O. N., DeCelles P. G., 2005. Structural geology and regional tectonic significance of the Ramgarh thrust, Himalayan fold-thrust belt of Nepal. Tectonics, 24, TC4008, 1-26, doi:10.1029/2003TC001617. Sapkota, S. N., Bollinger, L., Klinger, Y., Tapponnier, P., Gaudemer, Y., Tiwari, D., 2013. Primary surface rupture of the great Himalayan earthquakes of 1934 and 1255: Nature Geoscience, 6, 71-76.
Schelling, D., 1992. The tectonostratigraphy and structure of the eastern Nepal Himalaya: Tectonics, 11, 925-943. Schulte-Pelkum, V., Monsalve, G., Sheehan, A., Pandey, M. R., Sapkota, S., Bilham, R., Wu, F., 2005. Imaging the Indian subcontinent beneath the Himalaya: Nature, 435, 1222-1225. Sreejith, K.M., Sunil, P.S., Agarwal, R., Saji, A. P., Ramesh, D.S., Rajawat, A.S. 2016. Coseismic and early postseismic deformation due to the 25 April 2015, Mw 7.8 Gorkha, Nepal earthquake from InSAR and GPS measurements. Goephy. Res. Lett., doi:10.1002/2016GL067907 Upreti, B.N., 1999. An overview of the stratigraphy and teconics of the Nepal Himalaya: Journal of Asian Earth Science,17, 577-606. USGS, 2015. M 7.8 - 36 km E of Khudi, Nepal. http://earthquake.usgs.gov/earthquakes/ eventpage/us20002926 Wilson, D., Aster, R., Ni, J., Grand, S., West, M., Gao, W., Baldridge, W. S., Semken, S., 2005. Imaging the seismic structure of the crust and upper mantle beneath the Great Plains, Rio Grande Rift, and Colorado Plateau using receiver functions: J. Geophys. Res., 110, B05306, doi:10.1029/2004JB003492. Yagi, Y. and Okuwaki R., 2015. Integrated seismic source model of the 2015 Gorkha, Nepal, earthquake: Geophysical Research Letters, 42, doi:10.1002/2015GL064995.
Figure Captions: Figure 1: Locations of the 2015 Nepal Earthquake and its aftershocks overlain on the tectonic map of the region. The rectangular box represents the rupture zone of the earthquake. AB, CD, EF are the profiles along which the receiver function images are presented. Inverted triangles are the seismic stations used in this study. Relocated epicentres of the 1934 and 1833 earthquakes are also shown. Figure 2: Epicentral distribution of the earthquakes recorded by the a) HiCLIMB and b) HIMNT networks used in this study. Figure 3: The common conversion point image of receiver functions superimposed with the aftershocks and focal mechanisms of Mw > 5 events from the USGS catalogue along the profiles (a) AB, passing through the Mw 7.8 earthquake source region, (b) CD passing through Kathmandu, and (c) profile EF traversing the 12 May, 2015 Mw 7.3 aftershock/triggered event. AB and CD profiles use seismic data from the HiCLIMB network and EF profile from the HIMNT network. The colour coding suggests the change of seismic velocity. Figure 4: (a) The interpreted upper crustal structure and MHT configuration using the velocity structure and geological map pattern along AB profile. (b) The velocity structure along CD profile is overlaid with the crustal structure and MHT configuration after Pearson & DeCelles (2005). (c) The velocity structure along EF profile is overlaid with the crustal structure after Schelling (1992). Please note that the bulk of aftershocks occur below the MHT. Figure 5: (a) The combined coseismic ground deformation and modelled slip on the décollement during the 2015 Nepal (Mw 7.8 & 7.3) earthquakes after Lindsey et al. (2015). Please note the maximum uplift lies to the south of the peak slip on the décollement. (b) Coseismic vertical uplift during the Mw 7.8 event along CD profile and
(c) the composite uplift during Mw 7.8 and 7.3 events along the profile EF. Probable rupture is marked as green line along the sections. Figure 6: Reinterpreted crustal structures along (a) AB, (b) CD and (c) EF sections suggesting lateral variation incorporating a deeper splay/horse on the MHT forming a duplex zone encompassing the aftershocks distributions. Figure 7: Schematic section synthesizing the observations and structural interpretations of the 2015 earthquake and aftershock genesis. The lateral variation in upper crustal structure with duplex in the MHT towards the eastern margin and extent of rupture front on a blind thrust and associated ground deformation are indicated.
Figure 2a
90
60
30
Figure 2b
90
60
30
Graphical abstract
Highlights Correlation between the seismic structure and seismogenesis in the Nepal Himalaya. Delineation of a low velocity layer that corresponds to the Lesser Himalaya sedimentary duplex above the MHT flat . Lateral variations in upper crustal structure control rupture propagation Tectonic loading on the duplex caused internal deformation localizing bulk of the aftershocks.
S1367-9120(16)30450-3 http://dx.doi.org/10.1016/j.jseaes.2016.12.039 JAES 2908
To appear in:
Journal of Asian Earth Sciences
Received Date: Revised Date: Accepted Date:
23 November 2015 28 December 2016 28 December 2016
Please cite this article as: Pandey, A.K., Saikia, D., Ravi Kumar, M., Earthquake genesis in Nepal Himalaya: A perspective from imaging of the 25th April 2015 Mw 7.8 earthquake source zone, Journal of Asian Earth Sciences (2016), doi: http://dx.doi.org/10.1016/j.jseaes.2016.12.039
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.
Earthquake genesis in Nepal Himalaya: A perspective from imaging of the 25th April 2015 Mw 7.8 earthquake source zone Anand K. Pandeya, Dipankar Saikia b, M. Ravi Kumara,c* a CSIR-National Geophysical Research Institute, Hyderabad, India b Indian National Center for Oce4an Information Services, Hyderabad, India c Institute of Seismological Research, Gandhinagar, India ---* Corresponding Author: M. Ravi Kumar Email address: [email protected]
Abstract The Mw 7.8 earthquake in Central Nepal nucleated in the mid-crustal ramp zone of the Main Himalayan décollement Thrust (MHT) and propagated eastward for >140 km where the largest triggered event of Mw 7.3 occurred without any surface rupture. Although it is advocated that the slip and rupture dynamics are controlled by the structural configuration of the MHT and the upper crust, precise correlation between the seismic structure and seismogenesis is hitherto scarce in the Himalaya. To address the issue, we imaged the crustal structure along three profiles covering the earthquake source region using receiver function analysis of the seismic data from the HiCLIMB and HIMNT seismic networks to understand the lateral variability. A ~5 km thick, low velocity layer is observed at the mid–crustal level, that steepens in the MHT ramp zone. The bulk of the seismicity including large shocks after the 2015 Nepal earthquake lies in the vicinity of this low velocity layer. Correlation of the seismic structure and aftershock distribution with the published crustal structure clearly suggests that the rupture involves a thicker zone extending for >40 km to the south of the source zone in the MHT ramp. We refined the structure of the MHT zone incorporating published coseismic slip and ground deformation to suggest that the rupture terminated at the footwall imbricate (horse) on the floor thrust below the zone of maximum coseismic uplift and there was a two stage rupture towards the eastern margin of the rupture zone. Key words: Mw 7.8 Nepal Earthquake, Receiver functions, Crustal structure, Mid-crustal ramp 1. Introduction The 25 April 2015 earthquake of magnitude (Mw) 7.8 that occurred on a gently (710°) north dipping, near E–W trending thrust fault, at about 15 km depth (Fig. 1; USGS, 2015; Galetzka et al., 2015), is the largest event recorded in the Nepal Himalaya during the modern instrumental era. The rupture propagated eastward within ~40 seconds (Avouac et
al., 2015; Yagi and Okuwaki, 2015) where most of the aftershocks were clustered and the largest aftershock/triggered earthquake of Mw 7.3 occurred on 12th May, 2015 at a depth of 17 km. Over 500 local earthquakes / aftershocks with magnitude >4.0 occurred within 45 days of the main event (Fig. 1; USGS, 2015; Adhikari et al., 2015). It is construed that the hypocentre lies within the mid crustal ramp on the Main Himalayan Thrust (MHT) (Avouac et al., 2015), which is well established in this region (Ni and Barazangi, 1984; Schelling, 1992; Pearson and DeCelles, 2005; Avouac, 2015). The spatial distribution of the aftershocks defines a nearly 140 x 40 km2 rectangular fault zone that ruptured during the 2015 Nepal earthquakes (Fig. 1) abutting the presumed extension of the rupture zone of the 1934-M 8.4 Great Bihar Nepal earthquake (Adhikari et al., 2015). Results from InSAR line-of-sight displacement data gleaned from the ALOS-2 data show that the slip during the main shock and the largest aftershock/triggered event extends for over >140 km with a peak slip of 5.5 – 6.5 m on the MHT between 5 and 15 km depth and a >1m co-seismic uplift just north of Kathmandu and a corresponding depression on either side of the uplift zone (Lindsey et al., 2015; Avouac et al., 2015; Elliott et al., 2016; Sreejith et al., 2016). The main shock and the aftershocks of the 2015-Nepal earthquake did not produce any surface rupture and the observed ground deformation (Lindsey et al., 2015; Sreejith et al., 2016; Elliot et al., 2016) largely remains confined to the Lesser Himalayan region, which constitutes the locked zone of the MHT having an interseismic strain deficit rate of around 17.8 ± 1.5 mm/yr (Ader et al., 2012; Bilham et a., 1997) for over a decade in the region. The seismogenesis in the Himalayan region is controlled by the mid–crustal ramp on the MHT, which is the source zone of the earthquakes and microseismicity (Pandey et al., 1999; Schulte–Pelkum et al., 2005; Avouac, 2015), though its configuration is not well constrained. The large and great earthquakes rupture the southern flat of the MHT and emerge towards the Himalayan front depending on the co–seismic slip distribution and the geometry of the structures along the
rupture zone (Mugnier et al., 2013; Bollinger et al., 2014; Avouac, 2015). The 25th April, 2015 earthquake affected zone also experienced the Mw 7.6 – 1833 Nepal earthquake (Bilham, 1995; Mugnier et al., 2013) and 1934-Bihar Nepal Great earthquake, which produced surface rupture and ground deformation along the MFT in the adjoining region (Sapkota et al., 2013; Bollinger et al., 2014). In this paper, we performed common conversion point (CCP) receiver function analysis of the HiCLIMB (Nábělek et al., 2009) and HIMNT (Schulte-Pelkum et al., 2005) data along three sections AB- traversing the epicentre of the Mw 7.8 main event, CD- across Kathmandu, where maximum ground deformation is observed, and EF – across the epicentre of the largest aftershock/ triggered earthquake (Fig. 1). We correlated seismic receiver function analysis results with the crustal structures along the respective sections (Pearson & DeCelles, 2005; Schelling, 1992) and integrated them with the coseismic geodetic observations (Lindsey et al., 2015) to understand the seismogenesis and rupture during the 2015 Nepal earthquake. 2. Receiver function imaging of the crust The data accrued from the HiCLIMB and HIMNT networks offered an excellent opportunity to interrogate and validate the structure of the seismogenic zones in the Himalayan mountain belt using the receiver function (RF) analysis. Waveforms used in this study are extracted from the broadband stations deployed under the HiCLIMB and HIMNT seismological experiments in the Nepal Himalaya and southern Tibet and archived in the IRIS Data Management Centre. However, only a subset of data from stations of the HiCLIMB experiment, located within our study area south of 30ºN latitude is utilized (Fig. 1). Three component seismograms due to earthquakes with magnitude >5.5 in the epicentral distance range 30º to 90º having a P-wave SNR > 3 are selected for computation of receiver functions. A total of 1535 and 533 events that yielded good quality waveforms have a
reasonably good azimuthal coverage at the HiCLIMB and HIMNT stations, respectively (Fig. 2). Receiver function analysis is a powerful tool to investigate the receive side subsurface structure. The receiver function method is based on the fact that the coda of the Pwave contains the P to S (Ps) conversions and their reverberations resulting from the impedance contrast across layers beneath the receiver. These Ps conversions and reverberations can be isolated through deconvolution of the vertical component from the radial component of the seismic record of a teleseismic earthquake. However, the simple concept of deconvolution is tricky to implement due to instability arising from division by zero. To avoid this instability, several methods of receiver function calculation, like the spectral water level deconvolution (Langston, 1979; Owens et al., 1983; Ammon et al., 1990), deconvolution in the time domain by least-squares estimation (Abers et al., 1995), iterative time domain deconvolution (Ligorría and Ammon, 1999) and multitaper frequencydomain cross-correlation receiver function (MTRF) (Park and Levin, 2000), have been developed. The advantage of MTRF method over the other methods is the use of multitapers to minimize spectral leakage during estimation of the spectrum and use of a pre-event noise spectrum for frequency dependent damping. A drawback of this method is that it uses a short analysis window, which limits its usability in situations where information at longer lags is desired. In this study, the receiver functions are calculated using the extended-time multitaper frequency domain technique (EMTRF) of Helffrich (2006), which is an extension of the MRTF method of Park and Levin (2000) for computing receiver functions of arbitrary length. This method uses a series of short, overlapping, multiple tapers to window the time series across its length and sums the individual Fourier transformed signals to produce a receiver function estimate which preserves the phase information for each sub-window. The vertical and horizontal components of the seismograms are decomposed into P, Sv and Sh
components using the back azimuth and incidence angle estimated from the first 10 s of the seismograms around the P–wave arrival prior to applying the ETMTRF. A frequency domain low–pass cos2 taper with a cut off frequency at 1Hz is applied to the all the RFs to avoid Gibbs effect. After an initial quality check, 25,000 high quality RFs from 1848 events are retained for further processing. The common conversion point (CCP) stacking method has been used to image the structure along linear profiles shown in Fig. 1. The receiver functions are migrated to depth using Ps travel time equation and the ak135 velocity model whose crustal part is replaced by crust 1.0 (Laske et al., 2013) velocity model based on the location of the station. The depth migrated receiver function amplitudes are then projected onto a vertical plane along profiles AB, CD, EF covering the entire rupture zone of the 2015-Nepal earthquake (Fig. 1) and binned in 1x1 km grids using Fresnel zone approximation, summed and normalized by the total no RF’s in each grid (Fig. 3). The CCP image produced by this method along the profile CD (Fig. 3b) is slightly different from the that produced by Nábělek et al., (2009) in the same region. The difference is due to the fact that Nábělek et al., (2009) used a method of stacking migrated images produced using travel time equation for direct conversions (i.e Ps) and the multiples (i.e PpPs and PpSs+PsPs). Although, the stacking of migrated images for Ps and its multiples greatly enhances the Moho signal, the shallow structure may contain artefacts due to migration of Ps energy to shallow depths while migrating for the multiples (Wilson et al., 2005). Since the emphasis of this study is not on mapping the Moho discontinuity but on mapping the other shallow interfaces within the crust, our approach is more appropriate. Duputel et al., (2016) have also imaged the MHT zone along Hi-CLIMB network (C-D profile of present study) using P and S receiver functions computed using the iterative time domain deconvolution method (Ligorría and Ammon, 1999) and CCP stacking using a 1D velocity model of Grandin et al. (2015). Also, using a 1D velocity model in a region where large lateral variations in velocity structure are expected,
the CCP image produced by them may not necessarily represent the true shallow sub-surface picture in the main shock region. We try to overcome this issue by replacing the crust beneath each station with the relevant model from crust 1.0 and perform the CCP imaging. 3. Seismic structure and seismogenesis The CCP stacks are overlaid with the aftershock distribution and fault plain solutions of >M 5 events derived from USGS catalogue (Fig. 3). To understand the seismogenesis in Nepal, we interpreted the seismic images in conjunction with focal mechanisms of the Mw > 5 earthquakes that occurred during April–May, 2015 and the published crustal structures along sections CD (after Pearson & DeCelles, 2005) and EF (Schelling, 1992; Figs. 4). Since the seismic station coverage along section A-B passing through the main shock is sparse (Fig. 1), the crustal structure could not be imaged with a good resolution (Fig. 3a). The Mw 7.8 event occurred at ~12 km depth in a ~7-10° dipping MHT ramp zone (USGS; Galetzka et al., 2015; Adhikari et al., 2015) that is associated with a tabular low velocity layer, which has a diffuse but continuous southward extension along with the bulk of aftershocks in its vicinity (Fig. 3a). Based on map pattern and aftershock distribution, the generalized cross section with the extent of rupture is depicted (Fig. 4a). The dense network of HiCLIMB profile (Fig. 1) yielded a well constrained seismic section of the earthquake affected region passing through Kathmandu C-D (Fig. 3b). The CCP image reveals a crustal structure with higher velocities in the upper 7-10 km, underlain by a >5 km thick low velocity layer down to 10–15 km depth in the upper crust. Interestingly, most of the large aftershocks (Mw > 5) are confined to the low velocity layer in the Lesser Himalayan zone. This upper crustal low velocity layer was observed in the HiCLIMB section by Nábělek et al. (2009) but was not specifically discussed. The shallow high velocity structure can be correlated with the highly deformed Lesser Himalayan Crystallines and the underlying low velocity layer with the Lesser Himalayan sedimentary sequence as well as the highly sheared MHT zone (Fig. 4). This flat low velocity
slab deepens beneath the rising Himalayan topography to the south of the Main Central Thrust (MCT) zone, demarcating the ramp on the MHT where the large earthquake events have occurred (Fig. 3a, 5a) and the background seismicity is clustered (Pandey et al., 1999). Similar inferences were drawn in the Garhwal region where a narrow low velocity layer demarcates the MHT with a well defined ramp that coincides with the local seismicity observed during the 1999- Chamoli earthquake (Caldwell et al., 2013). To understand the structural control on seismogenesis, we overlay retro-deformable balance cross section by Pearson & DeCelles, (2005) over the seismic image (Fig. 4b). Unlike the Garhwal Himalaya, the tabular low velocity layer with internal perturbations along the CD section in Nepal does not correspond well to the MHT marked by Pearson and DeCelles, (2005; Fig. 4b). The low velocity zone is wider and a large number of aftershocks occur below the MHT flat (Fig. 4b). Further, the location of the ramp on the décollement and the Lesser Himalayan duplex in the ramp zone along section CD remain ambiguous and show a vague correlation with contrasting velocity layers (Fig. 4b). After the Nepal earthquake, Duputel et al. (2016) have also reprocessed the HiCLIMB data along C-D section identifying a simple MHT flat and the ramp at the base of low velocity layer where aftershocks are clustered. It was observed during the 2015 Nepal earthquake that the rupture propagated eastward and decelerated after 40 seconds (Avouac et al., 2015; Yagi and Okuwaki, 2015; Adhikari et al., 2015). The bulk of the aftershocks, including the largest aftershock/triggered earthquake of Mw 7.3 on 12th May 2015, are confined to the deceleration zone towards the eastern side of the rupture zone (Fig. 1) and the larger aftershocks are situated at deeper depths compared to those in the western margin. The seismic and crustal structures along the AB and CD sections (Figs. 4a, 4b) show a lateral variation. Since the mid-crustal ramp on the MHT is considered the earthquake source zone, the lateral variations in its geometry have profound influence on the seismogenesis. To understand the lateral variation in crustal
structure, we also generated a common conversion point image of receiver functions along profile E-F that samples the eastern side of the rupture zone (Fig. 1), using data from stations of HIMNT seismic network and overlaid the crustal structure and litho-tectonic disposition after Schelling (1992; Fig. 4c). Albeit having a low resolution owing to the sparse density of stations, the seismic section reveals crustal structure (Fig. 3c) that is distinctly different from that below the western segment (Figs. 3a, b) of the rupture zone. The projected aftershocks including the M 7.3 event are clustered in a relatively narrow zone in the vicinity of the low velocity layer but are located below the MHT and to the south of the ramp zone modelled by Schelling (1992; Fig. 4c). The earlier experiment along the HIMNT seismic section (~70 km east of the EF section) mapped the MHT as a diffused zone of seismic anisotropy and could not clearly image the mid crustal ramp on the MHT (Schulte-Pelkum et al., 2005). These observations warrant inclusion of a seismogenic low velocity layer and the zone of aftershocks in the Himalayan wedge to explain seismogenesis, in spite of the limitation due to sparse station coverage and poor seismic structure. 4. Coseismic slip and uplift pattern vis-a-vis lateral variability in the rupture The Mw 7.8 earthquake ruptured the mid-crustal MHT for more than 140 km, with the largest aftershock/triggered event of Mw 7.3 defining the eastern limit (USGS; Fig. 1). Results of InSAR interferometry from the earthquake affected region show coseismic ground deformation for over 100 km along a strike parallel narrow zone to the north of Kathmandu with maximum >1 m uplift and a corresponding depression towards north during the Mw 7.8 event (Fig. 5a, 5b) and ~50 cm uplift associated with the Mw 7.3 event (Fig. 5c; Lindsey et al., 2015; Avouac et al., 2015; Elliot et al., 2016; Sreejith et al., 2016). The modelled coseismic slip on the décollement suggests that the maximum slip of >5m on the mid-crustal MHT within the Lesser Himalaya and the maximum coseismic uplift lie to the south of the zone of maximum coseismic slip (Fig. 5a). The coseismic slip and uplift are clearly bi model
towards the eastern side of the rupture zone, where the Mw 7.3 event possibly released the residual strain with limited spatial extent (Fig. 1, 5). Interestingly, the aftershock pattern during the Nepal earthquake shows two zones of seismicity i.e. along the main event possibly in the source zone and another towards south beneath the Kathmandu basin (Adhikari et al., 2015). We compared the observed spatial extent of the coseismic uplift and the modelled coseismic slip on the décollement (Fig. 5; Lindsey et al., 2015) with the interpreted crustal structures along CD and EF sections (Fig. 6) and aftershock distribution pattern (Adhikar et al., 2015), to understand the rupture propagation. The multiple imbricates and the frontal horse emanating from a deeper (20-25 km depth) décollement (Pearson and DeCelles, 2005) do not conform with the modelled seismic structure and the aftershock distribution of the 2015 earthquake (Fig. 4b). We refined the mid-crustal structure incorporating the new observations (Fig. 6b). The steepening of low velocity slabs in the seismic section (Fig. 6b) beneath the Lesser Himalayan duplex, clearly define the ramp zone that flattens to the basal décollement at >15 km depth, possibly marking the southern extent of the aseismic segment of the MHT. The redefined MHT ramp connects to the southern flat, which encompasses the aftershock cluster and the velocity interface without disturbing the shallower structural configuration (Fig. 6b). This basal décollement together with the previously defined MHT flat (Pearson & DeCelles, 2005) forms a duplex zone encompassing most of the seismicity during the 2015 event. The aftershocks terminate at the zone of maximum co-seismic uplift, which possibly defines the frontal ramp/horse of the duplex structure on the basal décollement that ruptured during Mw 7.8 event (Fig. 6b). However, towards the eastern margin of the rupture zone, the MHT configuration (Schelling, 1992) does not confirm with the seismogenesis as the bulk of seismicity lies below the décollement (Fig. 4c). This zone experienced a two stage coseismic uplift of ~0.5 m each during the Mw 7.8 and Mw 7.3 events (Figs. 5a, 5c; Lindsey et al., 2015) possibly suggesting
multiple rupture during the respective events. The Mw 7.8 event produced ~0.5 m uplift corresponding to ~2- 3 m of modelled coseismic slip (Fig. 5a, 5c); whereas the Mw 7.3 event produced ~0.5 m uplift along the MHT ramp zone with >5 m peak coseismic slip (Fig. 5a; Lindsey et al., 2015) during the triggered rupture (Fig. 6c) accommodating slip deficit towards the eastern margin. We refined the MHT configuration primarily encompassing the aftershock distribution trend and the respective extent of ruptures during the Mw 7.8 and 7.3 events (Fig. 6c). This refined MHT configuration of the rupture zone along with that of Schelling (1992) forms a duplex in MHT zone with multiples imbricates / horses. Similar horse structure in the MHT (duplex) zone is proposed to explain the wider microseismic distribution and topography proxies in western Nepal (Harvey et al., 2015). The other unique structure along the EF section is the out-of-sequence Sun Kosi Thrust (SKT; Fig. 5b), which lies to the immediate south of the zone of maximum coseismic uplift (Fig. 1) and aftershock occurrence, without any noticeable co-seismic deformation during the 2015 Nepal earthquake. This suggests that the earthquake rupture terminates before the SKT imbricate on the MHT and probably within the MHT duplex at a greater depth than in the western flank (Fig. 6). We tried to analyse the coseismic slip and uplift distribution during the 2015 earthquake (Fig. 5) together with the interpreted MHT structure (Fig. 6). To accommodate bulk seismic slip during a thrust earthquake event without appreciable terrain uplift, it is imperative that the decollement thrust be wider, gently dipping, producing no significant throw (vertical) component during faulting (Fig. 6b, 6c) and/or the strain is accommodated by internal deformation within the duplex. The same is evident from the wider aftershocks distribution to the south of the MHT ramp (Fig. 6). However, the localized terrain uplift away from the peak coseismic slip (Fig. 5a, 5b) points towards steeply dipping thrust producing a larger throw (vertical component) and focussed uplift. Since, no surface rupture was observed
after the earthquake and no corresponding structure is exposed on the ground along the uplift zone, it possibly points to a blind thrust as an imbricate / horse on the MHT flat underneath the zone of maximum surface uplift causing linear surface uplift during rupture (Figs. 5, 6). Gehrels et al. (2003) have indicated front parallel thrusts within the Kathmandu nappe (Fig. 4b) and the out-of-sequence Sun Kosi Thrust (SKT; Figs. 1, 4c; Schelling (1992), to the east of the rupture zone which remains blind beneath the Kathmandu nappe; but no coseismic deformation is mapped or reported along these structures and their relation to the seismogenesis during the 2015 Nepal earthquake is not clear due to insufficient observations. 5. Discussion The Himalaya of Central Nepal has experienced several large earthquakes during the last millennium. Notable among them are the earthquakes in 1934, 1833, ~1255, and ~1100 AD, that severely affected the Kathmandu valley despite being at varied distance from the epicentres (Pandey and Molnar, 1988; Bilham, 1995; Mugnier et al., 2013; Sapkota et al., 2013; Bollinger et al., 2014). The rupture zone of the 2015 earthquake (Fig. 1; Avouac et al., 2015; Yagi and Okuwaki, 2015) overlies the relocated rupture zone of the Mw ~7.6 – 1833 earthquake (Bilham, 1995; Ambrasey and Douglas, 2004; Mugnier et al., 2013). The proximity of the relocated epicentre of the 1833 earthquake (Mugnier et al., 2013) to the largest aftershock (Mw 7.3) of the 2015 event suggests an interseismic interval of 182 years for the rupture on the same segment of the MHT with >5m of coseismic slip without any surface rupture. This suggests ~3.6 m of interseismic strain accumulation considering a ~20 mm/yr slip deficit on the locked segment of MHT to the south of the ramp (Bilham et al., 1997; Ader et al., 2012). The interseismic strain accumulation since 1833 along with the background strain and tectonic loading due to the great earthquake of 1934, which lies to the east of the present rupture zone, match closely with the relieved strain during the 2015-Nepal earthquake.
It is important to address the above ambiguity in the earthquake source zone and rupture in light of the seismic structure (Fig. 6) as the interseismic stress build-up triggers large earthquakes in the ramp zone and rupture the southern flat of the MHT. A comparison of the seismic structures, aftershock distribution pattern, published and interpreted crustal structures along the three profiles (Figs. 3,4,5,6) reveals lateral variation in the seismogenic structure of the 2015-Nepal earthquake zone, from west to east. The invariable association of the low velocity layer, albeit having a poor resolution along the EF section, suggests duplex structures in the MHT zone that accommodate strain during earthquake rupture propagation. However, this inference is under the assumption that the aftershocks are reasonably well located. This brings forth the focus on structural variations along the strike (lateral) in different segments of the Himalaya to explain varying recurrence intervals and rupture behaviour (Mugnier et al., 2013). Using a wider zone of seismicity and topography as proxies, it is argued that the region with a large flat or a wider duplex on the MHT in Lesser Himalayan zone experienced larger interseismic interval (Harvey et al., 2015). In contrast, the crustal structure in eastern Nepal is less complex with a large antiform growth over a sharp ramp on the MHT, which coincides with the zone of microseismicity and earthquake nucleation at mid crustal (~15-20 km) depths (Avouac, 2015). The hypocenters of the Mw 7.8 and 7.3 earthquakes were indeed located in this zone but the aftershocks have a wider distribution along the low velocity layer towards south of the ramp zone (Fig. 3). The low velocity is possibly caused by the lower density composition or higher fluid concentration in the rocks that retard the seismic wave velocity. The low velocity slab possibly represents the Lesser Himalayan sedimentary sequence primarily composed of limestone, sandstone, shale and quartzite (Schelling, 1992; Upreti, 1999). Further, the increasing complexity of mid-crustal structure, i.e. a simpler MHT ramp zone along AB section to the thickened duplex zone on the MHT towards the east (Fig. 4, 6c,
5), enhances the strength of the earthquake source zone that may be responsible for the increasing depth of large events from ~12 (Mw 7.8 event) to ~18 km (Mw 7.3 event) and the retardation of eastward rupture propagation after ~40-45s (Avouac et al., 2015; Yagi and Okuwaki, 2015). The lateral variations in structure may cause heterogeneity in the fault frictional behaviour affecting diminishing co-seismic ground deformation and rupture propagation towards east and lack of southward rupture during the Mw 7.3 aftershock (Fig. 6). 6. Conclusions We synthesize the salient observations and interpretations in a schematic diagram (Fig. 7). The decreasing thickness, increasing depth of the low velocity layer, increasing density and depth of aftershocks towards east and the interpreted structure on the seismic profiles (Figs. 3, 6) in the earthquake rupture zone suggest lateral variation in the upper crustal structure and the MHT configuration. The rupture propagated eastward from the locus of the Mw 7.8 earthquake and decelerated after ~40-45 seconds (Avouac et al., 2015; Yagi and Okuwaki, 2015) due to the duplex on the MHT, which strengthens the décollement zone towards east. The coseismic uplift of ~1 m (Lindsey et al., 2015) to the south of the zone of maximum (>5m) slip, along a strike parallel zone to the north of Kathmandu, marks the southern termination of rupture front on the blind thrust, which is possibly a steeper frontal ramp / horse on the MHT in the duplex zone, providing a weak zone for focused uplift during the 2015 earthquake. The same may be responsible for the repeated ground deformation and associated damage to the Kathmandu valley (Bilham, 1995; Pandey and Molnar, 1988; Ambraseys and Douglas, 2004; Mugnier et al., 2013; Sapkota et al., 2013; Bollinger et al., 2014) during past historical earthquakes. The eastern segment has produced composite rupture with the Mw 7.3 events filling the slip deficit in the rupture zone. Acknowledgements:
Authors acknowledge funding from the GENIAS and HEART programs of CSIRNGRI. IRIS-DMC is gratefully acknowledged for making the data available. We highly appreciate critical and constructive comments from guest editor J. P. Mugnier, Roland Burgmann and an anonymous reviewer that significantly improved the manuscript. References Abers, G.A., Hu, X. and Sykes, L.R., 1995. Sourch scaling of earthquakes in the Shumagin region, Alaska: time-domain inversions of regional waveforms. Geophysical J. Int., 123(1), 41-58. Ader, T., Avouac, J.P., Liu-Zeng, J., Lyon-Caen, H., Bollinger, L., Galetzka, J., Genrich, J., Thomas, M., Chanard, K., Sapkota, S., Rajaure, S., Shrestha, P., Ding, L., Flouzat, M., 2012. Convergence rate across the Nepal Himalaya and interseismic coupling on the Main Himalayan Thrust: Implications: Journal of Geophysical Research, 117, B04403, doi:10.1029/2011JB009071. Adhikari, L.B., Gautam, U.P., Koirala, B.P., Bhattarai, M., Kandel, T., Gupta, R.M., Timsina, C., Maharjan, N., Maharjan, K., Dahal, T., Hoste-Colomer, R., Cano, Y., Dandine, M., Guilhem, A., Merrer, S., Roudil, P., Bollinger, L., 2015. The aftershock sequence of the 2015 April 25 Gorkha–Nepal earthquake. Geophys. J. Int., 203, 2119–2124 Ambraseys, N. N., Douglas, J., 2004. Magnitude calibration of north Indian earthquakes: Geophys. J. Int., 159, 165-206. Ammon, C. J., Randall, G. E., Zandt, G., 1990. On the non-uniqueness of receiver function inversions. J. Geophys. Res., 95, 15,123–15,127. Avouac J.-P., 2015. Mountain Building: From Earthquakes to Geologic Deformation: In Gerald Schubert (Ed.) Treatise on Geophysics, 2nd ed., 6, Oxford: Elsevier, pp. 381-432.
Avouac, J.-P., Meng, L., Wei, S., Wang, T., Ampuero, J.-P., 2015. Lower edge of locked Main Himalayan Thrust unzipped by the 2015 Gorkha earthquake: Nature Geoscience, DOI: 10.1038/NGEO2518. Bilham,R., 1995. Location and magnitude of the 1833 Nepal earthquake and its relation to the rupture zones of contiguous great Himalayan earthquakes. Current Science, 69,101-128. Bilham, R., K. Larson, J. Freymueller, Project Idylhim members, 1997. GPS measurements of present day convergence across the Nepal Himalaya. Nature, 386, 61-64. Bollinger, L., S. N. Sapkota, P. Tapponnier, Y. Klinger, M. Rizza, J. Vander Woerd, D. R. Tiwari, R. Pandey, A. Bitri, S. Bes de Berc, 2014. Estimating the return times of great Himalayan earthquakes in eastern Nepal: Evidence from the Patu and Bardibas strands of the Main Frontal Thrust: Jour. Geophy. Res., 119, doi:10.1002/2014JB010970. Caldwell, W.B., Klemperer, S.L., Lawrence, J.F., Rai, S.S., Ashish, 2013. Characterizing the Main Himalayan Thrust in the Garhwal Himalaya, India with receiver function CCP stacking. Earth and Planetary Science Letters, 367, 15–27. Cattin, R., Avouac, J.P., 2000. Modeling mountain building and the seismic cycle in the Himalaya of Nepal: Journal of Geophysical Research, 105, 13389-13407. Duputel, Z., Vergne,J., Rivera L., Wittlinger, G., Farra, V., György H., 2016. The 2015 Gorkha earthquake: A large event illuminating the Main Himalayan Thrust fault. Geophy. Res. Lett., 43, 2517-2525. DOI: 10.1002/2016GL068083 Elliott, J.R., Jolivet, R., González, P.J., Avouac, J.-P., Hollingsworth, J., Searle, M.P., Stevens, V.L., 2016. Himalayan megathrust geometry and relation to topography revealed by the Gorkha earthquake. Nature Geoscience, 9, 174-180. DOI: 10.1038/NGEO2623. Galetzka, J., et al. (31 authors), 2015. Slip pulse and resonance of Kathmandu basin during the 2015 Mw 7.8 Gorkha earthquake, Nepal imaged with geodesy. Science, doi:10.1126/science.aac6383.
Gehrels, G.E., DeCelles, P.G., Martin, A., Ojha, T.P., Pinhassi, G., Upreti, B.N., 2003. Initiation of the Himalayan orogen as an early Palaeozoic thin skinned thrust belt. GSA Today, 13, 4–9. Grandin, R., M. Vallée, C. Satriano, R. Lacassin, Y. Klinger, M. Simoes, and L. Bollinger, 2015. Rupture process of the Mw = 7.9 2015 Gorkha earthquake (Nepal): Insights into Himalayan
megathrust
segmentation,
Geophys.
Res.
Lett.,
42,
8373–8382,
doi:10.1002/2015GL066044. Harvey, J. E., Burbank, D.W., Bookhagen, B. 2015. Along-strike changes in Himalayan thrust geometry: Topographic and tectonic discontinuities in western Nepal. Lithosphere, 7(5), 511–518. Helffrich, G., 2006. Extended-time multi-taper frequency domain cross correlation receiver function estimation. Bull. Seismol. Soc. Am.,. 96, 344-347. Jackson, M., Bilham, R., 1994. Constraints on Himalayan deformation inferred from vertical velocity fields in Nepal and Tibet: Jour. Geophy. Res., 375 99(B7), 13897-13912. Langston, C.A., 1979. Structure under Mount Rainier, Washington, inferred from teleseismic body waves. J. Geophys. Res., 84, 4749–4762. Laske, G., Masters., G., Ma, Z., Pasyanos, M., 2013. Update on CRUST1.0 -A 1-degree Global Model of Earth’s Crust: Geophys. Res. Abstracts 15, Abstract EGU2013-2658. Ligorría, J.P., Ammon, C.J., 1999. Iterat ive deconvolut ion and receiver function estimation. Bull. Seismol. Soc. Am., 89, 1395–1400 Lindsey, E.O., Natsuaki, R., Xu, X., Shimada, M., Hashimoto, M., Melgar, D., Sandwell, D. T., 2015. Line-of-sight displacement from ALOS-2 interferometry: Mw 7.8 Gorkha Earthquake and Mw 7.3 aftershock. Geophy. Res. Lett., 42, 6655-6661.
Mugnier, J. L., Gajurel, A., Huyghe, P., Jayangondaperumal, R., Jouanne, F., Upreti, B. N., 2013. Structural interpretation of the great earthquakes of the last millennium in the central Himalaya: Earth Science Review,127, 30-47. Nábělek, J., Hetnyi, G., Vergne, J., Sapkota, S., Kafle, B., Mei Jiang, Su, H., Chen,J., Huang, B.-S, the Hi-CLIMB Team, 2009. Underplating in the Himalaya-Tibet Collision Zone Revealed by the Hi-CLIMB Experiment. Science, 325, 1371-1374. Ni, J., Barazangi, M., 1984. Seismotectonics of the Himalayan collision zone: geometry of the under thrusting Indian plate beneath the Himalaya: Journal of Geophysical Research, 89, 1147-1163. Owens, T.J., Taylor, S. R., Zandt, G., 1983. Isolation and enhancement of the response of local seismic structure from teleseismic P-wave forms. Lawrence Livermore Lab. Report UCID 19809, 1-33. Pandey, M. R., Molnar P., 1988. The distribution of intensity of the Bihar-Nepal earthquake of 15 January 1934 and bounds on the extent of the rupture zone: J. Nepal Geological Society, 5(1), 22-44. Pandey, M.R., Tankudar, R.P. Avouac, J.-P., Vergne, J., Hritier T., 1999. Seismotectonics of the Nepal Himalaya from a local seismic network: Jour. Asian Earth Sci., 17, 703-712. Park, J., Levin, V., 2000. Receiver functions from multiple-taper spectral correlation estimates. Bull. Seismo. Soc Amer., 90, 1507-1520. Pearson, O. N., DeCelles P. G., 2005. Structural geology and regional tectonic significance of the Ramgarh thrust, Himalayan fold-thrust belt of Nepal. Tectonics, 24, TC4008, 1-26, doi:10.1029/2003TC001617. Sapkota, S. N., Bollinger, L., Klinger, Y., Tapponnier, P., Gaudemer, Y., Tiwari, D., 2013. Primary surface rupture of the great Himalayan earthquakes of 1934 and 1255: Nature Geoscience, 6, 71-76.
Schelling, D., 1992. The tectonostratigraphy and structure of the eastern Nepal Himalaya: Tectonics, 11, 925-943. Schulte-Pelkum, V., Monsalve, G., Sheehan, A., Pandey, M. R., Sapkota, S., Bilham, R., Wu, F., 2005. Imaging the Indian subcontinent beneath the Himalaya: Nature, 435, 1222-1225. Sreejith, K.M., Sunil, P.S., Agarwal, R., Saji, A. P., Ramesh, D.S., Rajawat, A.S. 2016. Coseismic and early postseismic deformation due to the 25 April 2015, Mw 7.8 Gorkha, Nepal earthquake from InSAR and GPS measurements. Goephy. Res. Lett., doi:10.1002/2016GL067907 Upreti, B.N., 1999. An overview of the stratigraphy and teconics of the Nepal Himalaya: Journal of Asian Earth Science,17, 577-606. USGS, 2015. M 7.8 - 36 km E of Khudi, Nepal. http://earthquake.usgs.gov/earthquakes/ eventpage/us20002926 Wilson, D., Aster, R., Ni, J., Grand, S., West, M., Gao, W., Baldridge, W. S., Semken, S., 2005. Imaging the seismic structure of the crust and upper mantle beneath the Great Plains, Rio Grande Rift, and Colorado Plateau using receiver functions: J. Geophys. Res., 110, B05306, doi:10.1029/2004JB003492. Yagi, Y. and Okuwaki R., 2015. Integrated seismic source model of the 2015 Gorkha, Nepal, earthquake: Geophysical Research Letters, 42, doi:10.1002/2015GL064995.
Figure Captions: Figure 1: Locations of the 2015 Nepal Earthquake and its aftershocks overlain on the tectonic map of the region. The rectangular box represents the rupture zone of the earthquake. AB, CD, EF are the profiles along which the receiver function images are presented. Inverted triangles are the seismic stations used in this study. Relocated epicentres of the 1934 and 1833 earthquakes are also shown. Figure 2: Epicentral distribution of the earthquakes recorded by the a) HiCLIMB and b) HIMNT networks used in this study. Figure 3: The common conversion point image of receiver functions superimposed with the aftershocks and focal mechanisms of Mw > 5 events from the USGS catalogue along the profiles (a) AB, passing through the Mw 7.8 earthquake source region, (b) CD passing through Kathmandu, and (c) profile EF traversing the 12 May, 2015 Mw 7.3 aftershock/triggered event. AB and CD profiles use seismic data from the HiCLIMB network and EF profile from the HIMNT network. The colour coding suggests the change of seismic velocity. Figure 4: (a) The interpreted upper crustal structure and MHT configuration using the velocity structure and geological map pattern along AB profile. (b) The velocity structure along CD profile is overlaid with the crustal structure and MHT configuration after Pearson & DeCelles (2005). (c) The velocity structure along EF profile is overlaid with the crustal structure after Schelling (1992). Please note that the bulk of aftershocks occur below the MHT. Figure 5: (a) The combined coseismic ground deformation and modelled slip on the décollement during the 2015 Nepal (Mw 7.8 & 7.3) earthquakes after Lindsey et al. (2015). Please note the maximum uplift lies to the south of the peak slip on the décollement. (b) Coseismic vertical uplift during the Mw 7.8 event along CD profile and
(c) the composite uplift during Mw 7.8 and 7.3 events along the profile EF. Probable rupture is marked as green line along the sections. Figure 6: Reinterpreted crustal structures along (a) AB, (b) CD and (c) EF sections suggesting lateral variation incorporating a deeper splay/horse on the MHT forming a duplex zone encompassing the aftershocks distributions. Figure 7: Schematic section synthesizing the observations and structural interpretations of the 2015 earthquake and aftershock genesis. The lateral variation in upper crustal structure with duplex in the MHT towards the eastern margin and extent of rupture front on a blind thrust and associated ground deformation are indicated.
Figure 2a
90
60
30
Figure 2b
90
60
30
Graphical abstract
Highlights Correlation between the seismic structure and seismogenesis in the Nepal Himalaya. Delineation of a low velocity layer that corresponds to the Lesser Himalaya sedimentary duplex above the MHT flat . Lateral variations in upper crustal structure control rupture propagation Tectonic loading on the duplex caused internal deformation localizing bulk of the aftershocks.