Revisiting the earthquake sources in the Himalaya: Perspectives on past seismicity

Tectonophysics 504 (2011) 75–88

Contents lists available at ScienceDirect

Tectonophysics j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t e c t o

Revisiting the earthquake sources in the Himalaya: Perspectives on past seismicity Kusala Rajendran ⁎, C.P. Rajendran Centre for Earth Sciences, Indian Institute of Science, Bangalore 560012, India

a r t i c l e

i n f o

Article history: Received 25 February 2010 Received in revised form 18 February 2011 Accepted 1 March 2011 Available online 5 March 2011 Keywords: Central Himalaya Earthquakes Paleoseismology Seismic gaps Northeast India

a b s t r a c t The ~ 2500 km-long Himalaya plate boundary experienced three great earthquakes during the past century, but none of them generated any surface rupture. The segments between the 1905–1934 and the 1897–1950 sources, known as the central and Assam seismic gaps respectively, have long been considered holding potential for future great earthquakes. This paper addresses two issues concerning earthquakes along the Himalaya plate boundary. One, the absence of surface rupture associated with the great earthquakes, vis-à-vis the purported large slip observed from paleoseismological investigations and two, the current understanding of the status of the seismic gaps in the Central Himalaya and Assam, in view of the paleoseismological and historical data being gathered. We suggest that the ruptures of earthquakes nucleating on the basal detachment are likely to be restricted by the crustal ramps and thus generate no surface ruptures, whereas those originating on the faults within the wedges promote upward propagation of rupture and displacement, as observed during the 2005 Kashmir earthquake, that showed a peak offset of 7 m. The occasional reactivation of these thrust systems within the duplex zone may also be responsible for the observed temporal and spatial clustering of earthquakes in the Himalaya. Observations presented in this paper suggest that the last major earthquake in the Central Himalaya occurred during AD 1119–1292, rather than in 1505, as suggested in some previous studies and thus the gap in the plate boundary events is real. As for the Northwestern Himalaya, seismically generated sedimentary features identified in the 1950 source region are generally younger than AD 1400 and evidence for older events is sketchy. The 1897 Shillong earthquake is not a décollement event and its predecessor is probably ~ 1000 years old. Compared to the Central Himalaya, the Assam Gap is a corridor of low seismicity between two tectonically independent seismogenic source zones that cannot be considered as a seismic gap in the conventional sense. © 2011 Elsevier B.V. All rights reserved.

1. Introduction The Himalaya, one of the most active interplate regions of the world has witnessed four great and many large earthquakes in recent history. Although cultural habitats and the tradition of documentation have existed in India for nearly four thousand years, records of ancient earthquakes are incomplete. The past two centuries have witnessed several significant earthquakes, four of which were of magnitude N8 (Fig. 1). Given such frequency of large/great earthquakes, it is reasonable to assume that the region might have experienced more earthquakes in the past, which are probably not documented. Even for the earthquakes identified from historic records, locations and magnitudes remain uncertain and there have been many recent attempts to reconstruct the earthquake history. For example, Iyengar et al. (1999) have used scriptures and other archival information as evidence for earthquakes in the medieval India, some of them dating to BC 2600–1800. Similarly, in the updated catalog of earthquakes of Northern India and Tibet, Ambraseys and Douglas (2004) have

⁎ Corresponding author. Tel.: +91 80 2293 2633; fax: +91 2360 0404. E-mail address: [email protected] (K. Rajendran). 0040-1951/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.tecto.2011.03.001

reevaluated the size and locations of earthquakes during the last 200 years; seven earthquakes (M N7) during AD 15–18 century are reported in their catalog. Interestingly, none of the great earthquakes of the past century has caused any primary surface rupture and thus the slip–magnitude scaling relations have to be inferred primarily from paleoseismological and geodetic data. Trenching excavations in the source zones of the past great earthquakes have exposed paleoliqefaction features, stratigraphic offsets and other evidence for past earthquakes (Kumar et al., 2001, 2006, 2010; Lave et al., 2005; Rajendran et al., 2004; Sukhija et al., 1999a,b, 2002; Wesnousky et al., 1999). Displacements recorded in some trenches on the foothills of the Central Himalaya have been interpreted to be of the order of 16–26 m (Kumar et al., 2006, 2010). Similar displacement (~17 m) has been reported also from trenches excavated near Nepal (Lave et al., 2005). Gaps of seismicity in time and space have been observed along the ~ 2500-km-long Himalaya plate boundary. The 500–800-km-long segment of the Himalayan arc between the rupture zones of the great 1934 Bihar–Nepal and 1905 Kangra earthquakes, known as the central seismic gap has been noted for its prolonged quiescence in terms of plate boundary events (Bilham et al., 2001; Khattri, 1987; Rajendran and Rajendran, 2005; Satyabala and Gupta, 1996, for example). Notable

76

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

Fig. 1. Map showing the Himalayan Frontal Thrust (HFT) and the significant earthquakes (source of data: Ambraseys and Douglas, 2004; Khattri, 1987; Oldham, 1883; Quittmeyer and Jacob, 1979).

among the large earthquakes in this region are the 1803, Mw ~7.7 and the 1833, Mw ~7.7 events (Ambraseys and Douglas, 2004; Bilham, 1995; Rajendran and Rajendran, 2005). The 1866, Mw ~7.2 and the 1916, Ms 7.1 are the other large earthquakes that occurred in this region during the last 200 years (Szeliga et al., 2010) Pacheco and Sykes, 1992). Earlier studies have indicated that both the 1833 and 1866 earthquakes have ruptured similar locations in the Nepal Himalaya (Khattri, 1987; Oldham, 1883). Study by Szeliga et al. (2010) based on the evaluation of intensity concurs with this and places this earthquake within 80 km of Kathmandu. The 1505 earthquake, with an estimated rupture length of ~600 km and moment magnitude Mw 8.2, is projected as a significant event to have occurred in the gap (Ambraseys and Douglas, 2004; Ambraseys and Jackson, 2003; Feldl and Bilham, 2006); although its status as a plate boundary event in the Central Himalaya is being debated (Rajendran and Rajendran, 2005). Despite these reported events, a Mw N8.0 earthquake is considered imminent in this region, based on plate convergence rates (Feldl and Bilham, 2006; Wallace et al., 2005). These observations call for a critical reevaluation of our current understanding of the past earthquakes in this region. Ever since the “Assam Seismic Gap” was proposed based on the estimate of rupture extent between the 1950 and 1897 events, it has been debated (Bapat, 1985; Khattri and Wyss, 1978). Both these earthquakes have been well documented and they remain two of the best studied, among the contemporary earthquakes (Ben-Menahem et al., 1974; Oldham, 1899; Rao, 1953). Paleoseismological investigations within their respective regions of strong ground shaking initiated during the last decade provide further understanding on the record of past events in this region (Kumar et al., 2010; Rajendran et al., 2004; Reddy et al., 2009; Sukhija et al., 1999a,b). In this paper we address two important issues concerning the seismicity of the Himalaya. One, we review the status of the central and Assam seismic ‘gaps’ using a variety of data that has already been gathered and two, we try to address the issue of the lack of any surface rupture from known great/large earthquakes of the last century as against the paleoslip observed in the recent trenches excavated in the foot hills of the Himalaya. 2. Tectonics and large earthquakes The Himalaya, ~2500 km-long and the highest mountain chain on Earth has resulted from the collision between India and Eurasia that started ~50–55 Myr ago with an estimated convergence of ~2000 to 3000 km (Molnar and Tapponier, 1975; Molnar et al., 1977). The collision has produced three major south-verging thrust faults that constitute the most significant tectonic features in the Himalaya, which

also accommodate the convergence across the Himalayan arc (Fig. 2A and B). The northern most and the oldest of this thrust system is the Main Central Thrust (MCT), which dips 30°–40° northward and marks the contact between the Higher and the Lesser Himalaya (Gansser, 1964). The age of the youngest deformation in the MCT zone is unknown (Hodges, 2000) and based on the absence of active deformation within the Quaternary deposits, Nakata (1989) argued that this thrust system is inactive. South of the MCT, the north-dipping thrust faults of the Main Boundary Thrust (MBT) separate the predominantly pre-Tertiary Lesser Himalayan sediments from the Tertiary and Quaternary sub-Himalayan sediments. The MBT is clearly expressed as a fault in the bedrock and locally it transports pre-Tertiary to Quaternary Lesser Himalayan and sub-Himalayan sediments over the younger Quaternary deposits (Nakata, 1989; Valdiya, 1992). The southernmost and the youngest of the three thrusts is the Himalayan Frontal Thrust (HFT), which is not as well exposed as the older ones, but occurs as a discontinuous range of scarps that cut the Quaternary fluvial terraces and alluvial fans (Kumar et al., 2001; Nakata, 1989; Valdiya, 1992). All of these south-verging thrusts within the Eurasian plate appear to merge with the Main Himalayan Thrust (MHT), the plane of detachment commonly referred to as the décollement (Hodges, 2000; Seeber and Armbruster, 1981) (Fig. 2B). Plate motion models and GPS measurements suggest that India– Eurasia convergence continues today at a rate of about 40 to 50 mm/year (Banerjee and Bürgmann, 2002). Thrusting on the HFT takes up about 10 to 20 mm/year of the total 40 to 50 mm/year of the convergence and the remaining motion is absorbed by a combination of thrusting, crustal extension and strike–slip motion within the Eurasian plate (Avouac and Tapponnier, 1993; Lave and Avouac, 2000). The continuing buildup of strain results in occasional great and large earthquakes along the Himalaya. The Himalaya plate boundary witnessed three major earthquakes during the past century and these are the 1905 Kangra (Ms 7.83); 1934 Bihar–Nepal (Ms 8.15), and 1950 Assam (Ms 8.48) earthquakes (magnitudes based on the revised estimates by Ambraseys and Douglas, 2004). The great plate boundary earthquakes are believed to have originated north of the MCT and ruptured the entire basal décollement south of the Higher Himalaya up to the HFT (Seeber and Armbruster, 1981). The surface geological data imply that the crust above the MHT is composed principally of thrust-imbricated sedimentary and crystalline strata that were detached from the leading edge of Indian plate (Gansser, 1964). Based on the analysis of great detachment earthquakes and focal parameters of medium size earthquakes, it has been suggested that a gently dipping (2–3°) regional detachment underlies the thrust faults at a depth of 15–20 km (Baranowski et al., 1984; Ni and Barazangi, 1984;

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

77

Fig. 2. (A) Major tectonic features of the Himalaya: MCT — Main Central Thrust, MBT — Main Boundary Thrust; ITS — Indus–Tsangpo Suture; SP — Shillong Plateau; ES — Eastern Syntaxis; AY — Arakan Yoma; Dark band — subrecent volcanics (base map modified from Ni and Barazangi, 1984; Nakata et al., 1990). Regions discussed in the paper are shown as shaded boxes. A — Central Himalaya; B — Shillong Plateau; and C — NE India. Inset: interplate thrust boundary between the Indian and Eurasian plates and the direction of ongoing convergence at the rate of 40–50 mm/yr (Banerjee and Bürgmann, 2002). (B) Generalized north–south geologic section showing various thrust systems in the Central Himalaya (after Seeber and Armbruster, 1981).

Seeber and Armbruster, 1981). This argument is consistent with the flattening of the Main Boundary Thrust at depth, as inferred from the balanced cross sections suggesting ramp geometry beneath the Lesser Himalaya (Célérier et al., 2009; Schelling and Arita, 1991; Srivastava and Mitra, 1994). Steepening of the Moho inferred from gravity anomalies also suggests an active ramp beneath the lesser Himalaya (Lyon-Caen and Molnar, 1985). Variations of terrace heights (300–500 m above the present river valley, with the heights of the lower terraces varying from 10 s of meters to 100 m) observed along the Kali Gandaki River, one of the major rivers of the Himalaya, may be pointed out as another evidence for the existence of the ramp structure (Molnar, 1987a). Further north, the mid-crustal ramp is believed to sole in to the deeper décollement that is considered to represent the 30–40 km deep seismic reflector imaged by the INDEPTH experiment (Zhao et al., 1993). A decade of geodetic and microseismic data from the Central Nepal suggest that the crustal ramps beneath the Higher Himalaya could be acting as a geometric asperity for the accumulation of elastic strain (Pandey et al., 1995). Further, these authors suggest that the transfer of stresses at the ramp-flat transition zones could evolve into large or medium size earthquakes. While most moderate/large earthquakes during the post-instrumental period seem to originate on gently dipping fault planes at focal depths of 10–20 km, some of them are also associated with sub-horizontal, north-dipping fault planes (Ni and Barazangi, 1984). Baranowski et al. (1984) reported three events with steeper planes (~25°) in the Western Nepal and its vicinity, which were considered to have originated above the detachment surface. The Mw 7.6, 2005 Kashmir earthquake sourced on a shallow and relatively steeper dipping fault may be considered as a recent example of ramp related, highly damaging earthquakes (Avouac et al.,

2006). The 1991 Uttarkashi and 1999 Chamoli, two of the betterstudied earthquakes of the last decade originated on gently dipping (~10°) fault planes, possibly the basal detachment. These earthquakes are discussed later in this paper. The south-directed over-thrusting from the Himalaya plate boundary and the subduction of the Indian plate beneath the Burmese plate along the Arakan Yoma have led to the complex tectonic structures of NE India (Fig. 2 A). The major seismotectonic elements are broadly demarcated as the Eastern syntaxis, Arakan Yoma subduction zone, Shillong Plateau and the Main Boundary Thrust (MBT) of the Himalaya (Dutta, 1964; Nandy, 2001). The 1950 earthquake is sourced near the eastern syntaxis, a tectonically complex region where east–west convergence is taking place within the Indo Burman ranges (Nandy, 2001). The other major earthquake in the vicinity of the Himalaya is the M 8.0, 1897 Shillong earthquake (revised magnitude; Ambraseys and Douglas, 2004). The source mechanism of this earthquake was attributed to a north-dipping detachment surface (Seeber and Armbruster, 1981), until Bilham and England (2001) proposed a south-dipping fault that bounds the Shillong Plateau as its causative structure (see Fig. 1 for location of the great earthquakes). Based on the current state of knowledge, the seismic sources are mostly attributed to the four tectonic elements mentioned above. 3. Status of the Central Seismic Gap The 500 to 800-km-long segment of the northwestern Himalaya, located between the rupture zones of the 1905 Kangra and the 1934 Bihar–Nepal earthquakes, generally referred to as the ‘Central Seismic Gap’ supposedly defines an unruptured part of the Himalayan arc

78

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

(Khattri, 1987; Khattri and Tyagi, 1983; Yeats and Thakur, 1998). Although this segment has remained relatively quiet since historic times, a few large earthquakes of M N 7.0 have occurred to its east, in the past. Of these, the earthquakes of 1833 and 1866 occurred in its eastern periphery, and the 1803 and the 1916 events occurred within the gap. Feldl and Bilham (2006) used the example of 1833–1934 pair during an interval of 101 years to suggest that a great earthquake can follow a major earthquake at the same location sooner than anticipated, based on the renewal times inferred from plate convergence rates, an apparent departure from the seismic gap theory. Following the same argument, they authors suggest that the 1905 Kangra region is ready for a large earthquake. Another event that has been discussed much in the recent years is the June 6, 1505 earthquake, which was originally believed to have sourced near northwest Nepal or southwest Tibet (Chitrakar and Pandey, 1986) until the recent evaluation of its size and location by Ambraseys and Jackson (2003). Based on the reported destruction to Tibetan monasteries placed ~700 km apart, a rupture length of 400– 700 km and Mw ~ 8.2 have been assigned to this earthquake, which is believed to have filled the entire central seismic gap (Ambraseys and Jackson, 2003; Bilham and Ambraseys, 2005). Based on the severity of damage in the Gangetic Plain caused by some of the previous earthquakes in the Central Himalaya (e.g., 1803 and 1934), it has been argued that a M N8 type earthquake would have caused much more devastation than what is reported for the 1505 earthquake. Moreover, the damage caused at locations separated by more than 600 km by some recent earthquakes (e.g., Mw 6.6, 1999 Chamoli; Rajendran et al., 2000) suggests that even moderate earthquakes in the Himalaya can affect sites that are separated by considerable distances (see also Martin and Szeliga (2010) for intensity reports for Indian earthquakes since 1636). Further, no destruction to the long-standing medieval monuments in the city of Delhi has been reported. Thus, we propose that there is a bias in using the damage to monasteries and temples, which are mostly located on hilltops, as a basis for evaluating the rupture extent for the 1505 earthquake (see Rajendran and Rajendran (2005) for further discussion on the 1505 earthquake). Geodetic data implies more than 9 m slip in the Central Gap, which can potentially generate a M ~8.7 earthquake and it has been suggested that if the recurrence interval is ~500 years, the region is ready for the next rupture (Bilham and Ambraseys, 2005). Based on paleoseismological investigations at various locations in the northwestern Himalaya, Kumar et al. (2006, 2010) reported a coseismic slip of 8–26 m that is ascribed to an event during AD 1400, which they believe could be the 1505 event. It is implicitly assumed that all these reported large vertical offsets in the geological sections are associated with the movement along the gently dipping detachment plane. We suggest that this paradox of 8–26 m coseismic slip reported from the sedimentary sections against the lack of slip from the great earthquakes in the recent past needs to be addressed. The historical uncertainties on the actual occurrence of great the late medieval great earthquake is another issue that needs to be understood in this context. 3.1. Major historic earthquakes in and around the Central Himalaya The September 1, 1803 event was regarded as the largest to have occurred in the central gap, until Ambraseys and Jackson (2003), reevaluated the reports on the 1505 earthquake and suggested it to be a M ~8.2, gap-filling event. Based on the damage reports, Rajendran and Rajendran (2005) placed the epicenter of the 1803 earthquake close to Srinagar (Fig. 3A, B) and assigned it a magnitude of Mw 7.7± 0.4. Some ancient temples in the region were damaged in 1803, the records of which are fortuitously preserved. For example, a stone inscription (in the ancient script of Devnagari) on the outer wall of the Gopinath temple at Gopeshwar testifies to its partial destruction and restoration after the 1803 earthquake (see Fig. 3B for location of the temple). The AD 10–12 century Katarmal Sun temple near Almora (appears to have been

Fig. 3. (A) Major structures in the Garhwal–Kumaun Himalaya with locations of the 1905, 1803, 1991 and 1999 earthquakes. (B) The traverses of the British explorer Raper (1810) around the epicentral area of the 1803 Garhwal earthquake and locations of temples and other features. The Uttarkashi and Chamoli earthquakes are shown by stars (after Rajendran and Rajendran, 2005). GN — Gopinath Temple at Gopeshwar; BN — Baijnath Temple; KT — Katarmal Temple.

affected by a shearing movement, as evident from the rotation on pillars that support two of the outer structures (see Rajendran and Rajendran, 2005 for details). During a visit to the temple in 1808, Raper (1810) had noted its ruinous state, but was uncertain if the damage was from the 1803 earthquake. During a recent visit to the region we noted some previously unnoticed evidence for rotation of pillars in the inner structure of the Baijnath temple, located about 5 km north of Almora. But for the minor damages on the three temples mentioned above, none of the other temples, whose histories are well known, is known to have suffered any shaking related damage. After detailed evaluation of the history and current status of the contemporary temples in the region, Rajendran and Rajendran (2005) concluded that the 1803 earthquake was not severe enough to destroy the large stone structures, the damage at Gopinath temple, close to the source, being the most significant. However, this earthquake damaged buildings in Delhi, located more than 250 km away from the epicenter (Nath et al., 1967). Notable among the damage to the distant city was at Qutb Minar, the 13 century 72.5-m high tower, which lost one of its cupolas (a plain square top on four stone pillars) (Cunningham, 1864; Sharma, 2001).

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

The June 6, 1505 event is reported to have strongly felt in the northern part of the Great Himalaya (Ambraseys and Jackson, 2003). However, as compared to the 1803 earthquake, there are fewer felt reports from the central Himalaya and the Gangetic Plain, for the 1505 event. This earthquake did not cause any significant damage to the monuments in Delhi. For example, the Bara-Gumbad Mosque (AD 1484; Sharma, 2001) apparently withstood this earthquake, indicating weak impact in Delhi. Trenching investigations in the NW Himalaya by Kumar et al. (2006, 2010) suggest a major slip event (coseismic slip ~ 18 m) during AD 1404 and 1422, which they have attributed to the 1505 event. Curiously, trenching by Lave et al. (2005) in the Nepal central Himalaya did not expose any slip corresponding to this supposedly gap-filling event, although they identified two older events (AD ~700 and 1100); the younger one was associated with a surface displacement of ~17 (+5/−3) meters. Slip associated with the 1505 earthquake was not observed in the trenches in Nepal Himalaya. Kumar et al. (2006) notes the lack of confirmed historical reports of large shaking commensurate with that expected of such a large event. Based on the relatively lower level of devastation in the Gangetic plain, Rajendran and Rajendran (2005) considered the 1505 earthquake as a non-plate boundary event, probably comparable to the 1803 earthquake. A few other recent, moderate earthquakes in the region have caused severe damage in their epicentral regions and far off locations — 1975 Kinnaur, 1991 Uttarkashi and 1999 Chamoli earthquakes, among them. The M 6.8 Kinnaur earthquake, reported to have been felt in an area of ~250,000 km2, including Delhi, caused wide spread destruction in the epicentral area (Khattri et al., 1978; Singh et al., 1975). Buildings in the affected areas were mostly of the mud/stone masonry type, which were severely damaged along with many temples and monasteries, usually located on hill tops (Singh et al., 1975), which curiously compares well with what have been reported for the 1505 event, near to its source zone. Although no primary surface rupture was recorded, Khattri et al. (1978) reported a series of step faults with near-vertical movement; the fault plane solution obtained by them (although not well-constrained) indicated a near-vertical maximum compressive stress, which they considered was quite unusual for a region dominated by NE–SW directed horizontal compression.

79

Himalayan earthquakes, and the rupture has not reached the surface. Yeats and Lillie (1991) related its source to a blind thrust, expressed on the surface by the growth of sub-Himalayan folds. The thrust faults and folds, caused by the highly deforming duplex zone observed in highest intensity zone of the 1905 earthquake (Fig. 4A and B), support this argument. The rupture presumably terminated near the mapped location of the Jawalamukhi thrust fault, located south of the Main Boundary Thrust (Powers et al., 1998). The overpressurization observed in exploratory wells and the interseismic strain reported by Gahalaut and Chander (1991), led Powers et al. (1998) to suggest that the subsidiary structures closer to the MBT may be subject to higher strain accumulation. Strain accumulation on the secondary structures associated with major thrust faults could result from the propagation of the fold-andthrust belt. The ramp geometry formed by the propagating thrust faults commonly lead to broad heterogeneities along the fault system, which act as geometric barriers for rupture termination and strain accumulation (Pandey et al., 1995, for example). Balanced cross sections in northwestern Himalaya (Powers et al., 1998; Srivastava and Mitra, 1994), and far eastern Nepal (Schelling and Arita, 1991) show the presence of crustal ramps overlying the great detachment plane. We have reproduced here the balanced cross sections across the

3.2. The 1905 Kangra earthquake The April 4, 1905 earthquake is among the three great 20th century earthquakes in the Himalaya, quite devastating in terms of the loss of life and property. Intensities of X (Rossi-Forel) were observed in the SubHimalaya at Kangra (Middlemiss, 1910). A magnitude Ms 8 was assigned to this earthquake using instrumental data (Gutenberg and Richter, 1954); based on the reevaluation using six European seismograms, Ambraseys and Bilham (2000) reassigned it a magnitude of 7.8 ± 0.2. No surface faulting was associated with this earthquake and in the absence of local seismometers, the aftershock locations are also not available to constrain the rupture. Although there have been efforts to use the triangulation data to model the rupture (Chander, 1988; Gahalaut and Chander, 1991; Yeats and Lillie, 1991), based on a reevaluation of the raw data, Bilham (2001) suggested that the triangulation suffers from several systematic errors and therefore rupture estimate remains questionable. The revised magnitude estimate of 7.8 ± 0.2 would suggest a rupture length of ~ 200 km and also rule out a continuous rupture from Kangra to Dehra Dun (Ambraseys and Bilham, 2000). The development of two high seismic intensity areas one close to the epicenter near Kangra and another near Dehra Dun about 250 km, to its southeast has remained a matter of interest. The double meizoseismal zones have been attributed to two independent earthquakes (Hough et al., 2005) as well as to secondary effects from site response on thick alluvium (Srivastava et al., 2010). Focal depth of this earthquake is not reliably determined. Seeber and Armbruster (1981) suggested that the 1905 Kangra earthquake also originated on the basal décollement, like the other great

Fig. 4. Views of the liquefied sections observed near Gagil. (A) Close view of a sandblow feature (dotted outline) wherein fluidized fine sand has broken through the top brownish clayey–sandy layer. The location of the peat sample is shown by a dot. (B) sections of fluidized sand cutting through pebble horizon, and (C) fluidization below a thick layer of gravel.

80

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

Kangra region, showing development of crustal ramps and steep fault planes on the thrust systems (Fig. 5A and B; Powers et al., 1998). Given this geometry, the thrust sheets must bend upward as displacements are translated up a ramp, causing near-surface uplift. Vertical deformation of the thrust sheet is also believed to be controlled by the slope of the fault as well as volume contraction within the wedge, leading to local uplifts, which may not be associated with coseismic slip (Jackson and Bilham, 1994). Arguably, the asperity breaking earthquake, originating on the crustal ramp can transfer movement along their steep limbs, leading to large displacements compared to those originating on the detachment plane. A similar mechanism has been evoked to explain the coseismic surface slip up to 10 m, associated with the Mw 7.6, 1999 Chi-Chi (Taiwan) earthquake (Chen et al., 2001). The Mw 7.6 2005, Kashmir earthquake, comparable to the Chi-Chi earthquake, was associated with an average fault slip of 4 m, with a maximum of 7 m and the surface ruptures are reported to coincide with the MBT (Avouac et al., 2006). The Kashmir earthquake is comparable in size to the Uttarkashi and Chamoli events, but it was far more destructive, probably because of its ground deformation characteristics. The Uttarkashi and Chamoli earthquakes were caused by the rupture of shallow dipping (~10°N) blind thrust faults on deeper parts (~15 km) of the detachment (Cotton et al., 1996; Kayal et al., 2003). On the contrary, the Kashmir earthquake occurred on a relatively steep fault that splays upward from the detachment, as suggested by the focal mechanisms (Harvard CMT: strike of fault plane: N133°E, dip 29°; Avouac et al. (2006). 3.3. Result of Trenching Investigations around Almora The regions affected by the 1803 earthquake are considered favorable locations for trenching investigations and we selected sites close to the Katarmal temple. Here we report soft sediment deformations observed near Gagil, located about 8 km north of Almora, on the east bank of the Kosi River (Fig. 3B for location). We examined a nearly

10 m-long and 8 m-high section at the base of which we found a fluidized layer made up of yellow and grayish clayey sand. Vertical intrusions composed of whitish fine sand have disturbed the overlying layer of grayish clayey sand (Fig. 4A). These sand-filled, fissure-like structures are similar to what are classified as “pillars” that originate at the bases of sand units overlying layers of mud or clay (Lowe, 1975). Such upward movement of sand usually results from forceful ejection of water during an earthquake shaking. Signs of forceful flow were evident also from the rip-up clasts present within the sand. Peat embedded within the clayey–sandy layer, which was broken during the ejection of sand, yielded an age of 808 ± 60 years BP (AD 1119–1292). This could represent the age of the layer disrupted during the ejection of sand or the peat could have been remobilized during the fluidization. Either way, this date may represent maximum age of the fluidization event. In another section ~3 km east of the site at Gagil, fluidized sand has been mobilized through thick overburden, consisting mostly of pebble– gravel mix, the fluidized layer currently occurs ~10 m below the surface (Fig. 4B). At another site ~1 km north of the Gagil site, fluidized sand has been mobilized through the overlying pebble bed, but no datable material was obtained from this layer (Fig. 4c). Evidence for mobilization of fine sand through the overlying strata was observed also at a few other nearby sites, suggesting that fluidization is not isolated. Based on the rotation of pillars of the ancillary structures of the AD 10–12 century Katarmal temple and the evidence of fluidization at multiple sites in its vicinity, we suggest that an earthquake may have affected this region after this temple was constructed. The age obtained for the feature at Gagil falls in the same range (AD 1100) as reported by Lave et al. (2005) from the trenches in the Central Nepal Himalaya. 4. Status of the Assam Gap The northeastern region of India has generated two great earthquakes (1897 and 1950, both M N8). The 1897 earthquake occurred on the Shillong Plateau, and the 1950 earthquake occurred on the Assam

Fig. 5. (A) Duplex zone around the rupture termination of the 1905 Kangra earthquake. MCT — Main Central Thrust; MBT — Main boundary thrust. (B) Balanced Cross section along AA-A as shown in Fig. 5A. (Modified from Powers et al., 1998).

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

syntaxis, the northeastern part of the Indian plate (Fig. 6A, B; see also Bhattacharya et al., 2005; Gowd et al., 1998; Kayal and De, 1991; Thingbaijam et al., 2008 for reviews on the seismicity of Northeast

81

India). A conspicuous feature of the northeast India seismicity is the ~250 km-long-spatial gap between the ruptures of 1897 and 1950, designated as the “Assam Gap”, (Khattri and Wyss, 1978; Khattri et al.,

Fig. 6. (A) Map showing the important tectonic features in NE India. (B) Relief map of the NE India showing the epicenters of significant earthquakes in the NE India (Source of data: Ambraseys and Douglas, 2004; Ambraseys and Jackson, 2003; Iyengar et al., 1999; Yadav et al., 2009).

82

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

1983). Seeber and Armbruster (1981) noted that the 1897 earthquake rupture was 550-km-long, filling the ruptures between 1934 and 1950 earthquakes, consequently lowering the potential for a large earthquake in this segment of the Himalaya. Based on the geometry of the source zones and the conventional definition of seismic gaps, Gowd et al. (1998) argued that the suggested “Assam Seismic Gap” is not a real gap in seismicity. Kayal and De (1991) also refer to this elongated zone as an “aseismic corridor”, implying that large earthquakes are unlikely in this region. A more recent study by Thingbaijam et al. (2008) also suggested a northeast trending zone of low seismicity between the Mikir Hills and the Mishmi thrust. Thus, the idea of “Assam Gap” as a zone of future great earthquake remains to be validated, possibly through extensive paleoseismologic investigations to identify past ruptures and sources located within this region. In the following sub section we provide an update on the past histories of these two and other sources identified in the Assam valley. 4.1. The 1897 Shillong earthquake The 1897 earthquake is one of the most damaging and perhaps the best-documented great earthquake during the 19th century (Oldham, 1899). The earthquake produced an ~500-km-long and possibly 300km-wide rupture and is believed to have originated on a northdipping fault that crops on the Shillong Plateau (e.g. Molnar, 1987b; Seeber and Armbruster, 1981; Fig. 7A). This was the prevailing view until Bilham and England (2001) proposed a steep (50°) south– southeast dipping reverse fault, close to the northern topographic front of the Shillong Plateau. Based on geological, geophysical and seismological data, Rajendran et al. (2004) suggested a less steep, south-dipping fault, further north of the Oldham Fault, which would project under the Brahmaputra alluvium, an issue that remains debated (Bilham, 2006; Nayak et al., 2008; Rajendran et al., 2006a). The other issue concerns the timing of the past earthquake in the region. Based on historical, archeological and paleoseismological data, Rajendran et al. (2004) suggested that the penultimate earthquake in the 1897 source may have occurred ~1000 years BP, whereas Sukhija et al. (1999a,b) suggested a date closer 500 yr BP. These issues have also been debated (Rajendran et al., 2006b; Sukhija et al., 2006). We briefly review the previous studies and present more information on the past seismicity of the region.

4.1.1. Penultimate event: evidence from paleoseismogical and archeological data Evidence for a pre-1897 event, dated between AD 790 and 1010 was first reported from Mendipathar (Dilma), close to the Chedrang fault (Rastogi et al., 1993; see Fig. 7B for locations of trenches). They further reported that sand dikes belonging to the same vintage observed here are ≥50 cm in width and are ubiquitous in this region. Later studies at other close by spaced locations reported evidence for two past earthquakes which were dated at 500 ± 150 yr BP; 1100 ± 150 yr BP and possibly a third event N1500± 150 yr BP (Sukhija et al., 1999a,b). The oldest event was based on a single feature where as the other two were averaged from multiple dates. Rajendran et al. (2004) reported multiple liquefactions from trenches excavated near Dilma, and suggested maximum and minimum age limits between AD 645–980 (1250 ± 80 yr BP) and BC 545–AD 129 (2170 ± 140 yr BP). In another section at Dilma, located a few meters west of the above mentioned site, peat from a ruptured clay horizon overlying a layer of fluidized sand yielded an age of 2501 ± 75 yr BP (BC 810–400; see Table 1). They also note that the younger features (AD 1450–1650) are characterized by ~10 cm wide vents. The older vents (AD 700–1050) are ≥50 cm wide and are comparable to the 1897 features, as reported also by Rastogi et al. (1993). Historical and archeological data provide additional constraints on the past earthquakes that affected the Brahmaputra valley, where a rich cultural establishment flourished during AD 5–10 century. Widespread destruction to stone temples around Dhubri, Goalpara, Barpeta, Guwahati, Nowgong and Tezpur, all belonging to AD 9–10th century has been reported (Chaudhury, 1964a,b; see Fig. 7A for locations). Barua (1966) argued that only some major earthquakes could have caused such widespread destruction and that the period between AD 9–11 centuries was known for massive reconstruction activity. The case of a sixth century structure, the ruins of one of the oldest temples in Assam, is particularly instructive. Located at DaParbatia, near Tezpur, the remains of the original structure dates back to AD 5–6th century, which was probably destroyed during an earlier earthquake (Chaudhury, 1964a,b). Only a single doorframe remained from the earlier construction and during AD 18 century, a new structure was constructed around this frame. This new structure collapsed during the earthquake of 1897, revealing the doorframe of the older structure (Barua, 1966, Fig. 7C). Coexistence of structures

Fig. 7. (A) Map showing the Brahmaputra Fault and the epicenter of the 1897 earthquake. (B) Map showing the locations of trenches near Chedrang Fault (C) Remains of the temple at Da Parbatiya. (D) Sketch of the Barnadi Bridge, which collapsed during the 1897 earthquake (sketch by Hannay, 1851).

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

83

Table 1 Radiocarbon and TL dates from various sites in NE India. Location

Sample ID/Ref

Lab code

14

Calibrated age (years AD)

Sample/comments

Dilma - do- doBeltaghat Kharidhara Beltaghat Medipathar Kamakhya - doKalikulla Kalikulla Kalikulla Kalikulla Paya

D1-1897-1 D2-1897-1 D2-1897-2 SP-11 Sukhija et al. (1999b) SP-28 Sukhija et al. (1999b) SP-16 Sukhija et al. (1999b) Rastogi et al. (1993) KM/S2/BR KM/S3/Br B1/D/E1/1 B1/D/E1/2 B1/D/E1/4 B1/D/E1/3 P1/D/E1/4

NZAa14919 BSb-1900 BS-1901

2501 ± 75 2170 ± 140 1250 ± 80 1010 ± 150 440 ± 140 1500 ± 150 1110 ± 100 1194 ± 120 yr 1464 ± 150 yr 1460 ± 220 yr 1910 ± 690 yr 200 ± 80 yr 5040 ± 110 yr 80 ± 90 yr

BC 800–400 BC 545–AD129 AD 645–980 AD 870–1190 AD 1380–1620 AD 395–660 AD 790–1010 682–926 521–686 AD 335–781 BC 776–AD 713 AD 1797–1956 BC 4053–3634 AD 1794–1956

Charcoal Charcoal Charcoal Wood Charcoal from sand dyke Peat from sand blow Tree trunk bricks (layer 2) Bricks (bottom layer) Twig Peat-rich sediment Peat In situ tree trunk Peat-rich sediment

KS2c KS3 BS-1722 S-2432 BS-1728 S-2438 BS-1736 S-2477 BS-1727 S-2437 BS-1738 S-2449

C age (yr. BP)

a

: NZA: analyzed at Accelerator Mass Spectrometry (AMS) laboratory at Rafter, New Zealand. : BS: analyzed at Radio Carbon Dating Laboratory, Birbal Sahni Institute of Palaeobotany, India. Calibrated using CALIB Radiocarbon Calibration (version 5.0.1), Stuiver et al. (1998). c : Analyzed at the Thermolumniscence laboratory at the Imphal University, Manipur, India. b

that follow distinctly different styles of construction and materials used could have resulted from the restoration of existing structures, following contemporary styles and building materials. Such ancient ruins are not just confined to the vicinity of Tezpur and Guwahati, but they are also found at many other locations in the Brahmaputra Valley (Banerji, 1923; Chaudhury, 1964a,b), suggesting that the destruction was rather wide spread. Some ancient books such as Kalikapurana and Yoginitantra (AD 9 and 16 century respectively, referred by Iyengar et al., 1999) make references to destruction to the temple of Kamakhya (AD 7–8th century) situated on the top of Nilanchal Hills (an inselberg of the Shillong plateau), about 8 km southwest of the town of Guwahati (Fig. 7A for location). The excavations (~2 m deep) at the temple of Kamakhya exposed a layer of debris, broken bricks, pottery and a jumble of carved stones with alternating levels of bricks, confirming the historical records of destruction (see Rajendran et al., 2004 for details). Pottery and the bricks from the bottommost layer yielded thermoluminescence (TL) dates of 1194 ± 120 yr and the 1464 ± 150 yr. Rajendran et al. (2004) interpreted this as the minimum and maximum age limits for the penultimate earthquake. The younger age obtained from pottery is closer to the age of the penultimate earthquake in the 1897 source, inferred from trenches excavated at Dilma. Another important structure that testifies structural integrity until 1897 is a 5 to 10 century-bridge across the presently dry channels of the Barnadi River close to Guwahati. We refer to this structure as the Barnadi Bridge, which was functional at least until 1205–1206 as reported by Hannay (1851) (Fig. 7D). This bridge collapsed during the 1897 earthquake (Barua, 1966; Gait, 1905; p. 141). Since the bridge was usable during AD 1205–1206 and it collapsed only in 1897, we surmise that this bridge had probably not experienced similar shaking at least for about 700 years of its existence. 4.2. The 1950 Assam earthquake The August 15, 1950 earthquake in Upper Assam close to the northern boundary of eastern syntaxis is one of the rare historic earthquakes, which has been instrumentally recorded. The earthquake was located at Rima, near the Indo-China border (Tandon, 1954); it has been assigned a magnitude of 8.6–8.7 (Richter, 1958). This earthquake was felt over an estimated area of 3 million km2 in India, Burma, East Pakistan, Tibet and China and caused extensive damage to Sadiya, Saikhoaghat and many other populated towns of Assam, leading to 1500 deaths. It was most damaging in the Upper Assam and produced remarkable changes in the topography of the Upper Brahmaputra valley. The riverbeds were heavily silted, leading to floods and shifts in the course of rivers. Damage to roads, railways,

bridges and buildings were extensive and the estimated acceleration on alluvium was 0.5 g (Ben-Menahem et al., 1974). Significant changes caused by landslides were observed in the landscape, drainage and ecology of an alluvial tract exceeding 150 km2 on the north bank of the tributaries of the Lohit (alias Upper Brahmaputra) River. Catastrophic destruction of extensive forests was evidenced by numerous dead tree stumps in an upright position embedded in the sand sheets along river courses (Kingdon-Ward, 1951; Ramesh and Gadagkar, 1990). The aftershocks originally located by Tandon (1954) and relocated by Chen and Molnar (1977) were distributed over a wide zone that suggest a complex pattern of rupture, which was interpreted as a consequence of excitation of subsidiary faults (Ben-Menahem et al., 1974). The focal mechanism of the main event obtained by the same authors suggested a mechanism dominated by strike–slip faulting. The two suggested planes were in the NNW–SSE and ESE–WSW directions, the former supported by the regional structural trend (Mishmi and Lohit Thrusts) as well as distribution of initial aftershocks. Chen and Molnar (1977) suggested thrust faulting on an easterly striking, northward dipping fault at the eastern end of the Himalayan plate boundary and strike–slip motion along the adjacent NNW striking plate boundary in Burma. Focal mechanisms of more recent earthquakes reported by Chen and Molnar (1977) also show large component of thrust faulting along and near the eastern end of the Himalaya plate boundary, suggesting that a combination of strike slip and thrust mechanisms is possible for the 1950 earthquake. 4.2.1. Historical seismicity Documented records do not mention any event of comparable dimensions in the source zone of the 1950 earthquake. However, significant earthquakes have been reported from the Assam valley and nearby areas since the AD 15 century (see catalogues by Ambraseys and Douglas, 2004; Iyengar et al., 1999). Some of the better-known historical earthquakes in the region are shown in Fig. 6. Here we discuss some of the important events discussed in the literature. 4.2.1.1. 1697 Sadiya earthquake. The 1967 Sadiya earthquake is one of the earliest and most damaging earthquakes in this area, felt strongly at the town of Sadiya, ~ 150 km southwest of the 1950 source. The court chronicles mention of an earthquake in AD 1697, which occurred at Sadiya in the Upper Assam region (referred to by Iyengar et al., 1999) which reads as follows: In the month of Puh 1618 (AD 1697) Bandar Phukan of the Chetia family constructed a fort at Puingdang under the orders of the king which took two months. In the same year there was an earthquake, which continued for six months in an abortive fashion, from Phagu to Saon. The earth was rent asunder at Sadiya, and magur and kawai fish appeared in the breaches. As sands and

84

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

water appeared at that place, the sides of the hills crumbled down. This is an explicit description of a major earthquake, followed by a long and significant sequence of aftershocks. The earthquake is reported to have caused an appreciable aftershock sequence and massive damage. Ground fissuring and liquefaction were reported from the regions close to Sadiya. The endurance of the Barnadi Bridge discussed earlier suggests that the 1697 event must have been quite far from the site of the bridge, and also from the 1897 source. The damage intensity at Sadiya during the 1950 earthquake as indicated in Poddar (1950) appears to be similar to what is described for the 1697 earthquake and thus we believe that the event was closer to the 1950 source. 4.2.1.2. Earthquake of 1713/1714. A destructive earthquake is reported to have occurred in the northeast India during the spring of 1713 and it destroyed many houses and caused many fatalities. This event is also mentioned in numerous contemporary Butanese documents, although there is lack of clarity on the year of its occurrence (Ambraseys and Jackson, 2003). The court chronicles describe this event as follows: The stone domes situated at the two doors of Bhagra gate were broken by some mysterious agent at night. The domes of the temple at Tinkhang were also broken as well as those at the subsidiary temple at Charaideo Hill Iyengar et al., 1999). Charaideo Hill is located to the southeast of Sibsagar (26.98°N, 94.5° E), as reported by Gait (1905). This earthquake appears to be what Iyengar et al. (1999) describe as the AD 1714 event, going by the descriptions of damage to the temple at Charaideo Hills. Although its exact location is uncertain, this event may have occurred to the north of the Mikir Hills (Fig. 6B). Locations of a few other moderate to large events mentioned in the literature are also shown in Fig. 6B, of which the 1548 earthquake is notable as the recent work by Reddy et al. (2009) who identified geological evidence for this event. 4.3. Results of trenching excavations around the Lohit River region The 1950 earthquake generated widespread fissures and sand vents in many localities in the alluvial plain of the Lohit River. The worst affected areas were in the vicinity of Saikhawaghat, Sadiya, Margherita, Ledo and Jorhat. In many places the fissures and sand vents are reported

to follow a linear arrangement and in some cases this continued for miles. Most of these fissures and sand vents are reported to be parallel to the rivers and abandoned streams (Mathur, 1953). We carried out some exploratory work around Saikhawaghat, concentrating mostly on the northern and southern banks of the nearly east–west flowing Lohit River (Fig. 8), and the preliminary results were discussed by Rajendran and Rajendran (2000). Trenching at selected locations, Digaru Ghat, Paya and Saikhawaghat in Lohit district exposed evidence of massive liquefaction, mostly in the form of vented sands (Fig. 9A and B). At some sites we located large size vents of recent origin possibly due to the 1950 earthquake. Reddy et al. (2009) have carried out trenching excavations in the area and the ages obtained by them mostly correspond to the 1950 event. They further provide evidence for two historically reported earthquakes (AD 1548 and 1697). Our excavations at Digaru Ghat and Paya revealed a complex and reworked stratigraphy, with evidence of sand blows. While the evidence for relatively younger events are better preserved, the older features appear to have been highly disturbed (possibly due to the frequent floods, erosion and landslides that frequent this area). Here we discuss one of the sections exposed at Digaru Ghat, where the 3 × 2 m trench exposed essentially medium to fine grained gray sand, which has been intensely liquefied. The trench exposed three units of fine-grained whitish sand, which we refer to as the top, middle and bottom layers of liquefied sand (Fig. 9A,B). The host sand is composed of silty, greyish sand. Rip-up clasts derived from the host sands observed in the liquefied unit suggest forceful ejection of water. The top layer of white sand showing signs of fluidization probably represents the youngest event, the 1950 earthquake. The liquefied layer below this event horizon is capped by a thin layer of peaty material (2–3 cm), which probably represents the paleo surface. The peaty material was dated at 200 ± 80 yr BP. (calibrated age: AD 1797 to Present), which provides the maximum age of the intermediate unit. We suspect that this mid-layer of liquefied sand could have been generated by the 1697 Sadiya earthquake that caused significant liquefaction in the Brahmaputra Valley. Although the exact location of the 1697 earthquake is not known, maximum damage was reported from Sadiya, located about 50 km NW of the trench at Paya. The

Fig. 8. Map showing the epicenter of the 1950 earthquake and the affected area. Hatched area shows areas affected by the landslides (modified from Mathur, 1953). F: location of the damaged fort at Chimri; T1 and T2: Exploratory trenches, which exposed the liquefaction features.

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

85

Fig. 9. (A) Photograph showing the liquefaction features and part of a tree stump exposed in trench T2 at Digaru Ghat. (B) Sketch of the liquefaction features exposed at Trench T2. Dots show sample locations.

bottom-most layer of liquefied sand also consisted of the same source sand, although we could not trace the fissure for any of these features, in this trench. A twig obtained from this layer gave an age of 1460± 220 yr BP (calibrated age: AD 335–781) which we consider as the contemporaneous date of the sand emplacement. A tree trunk in its growth position was observed at ~1.6 m below the present surface, which was dated at 5040 ± 110 yr BP (BC 4053–3634). From the disposition of the trunk, it appears that tree was ripped off in a sudden movement. The 1950 earthquake is known to have caused extensive destruction to the forests in the region, as evidenced by the numerous tree stumps in upright position (Ramesh and Gadagkar, 1990). 5. Discussion The Himalaya plate boundary and the Shillong plateau have generated four great earthquakes during the 100 plus years, but the record of seismic history before AD 1500 remains sketchy. Newer tools and techniques propelled studies during the last two decades — GPS based slip models and paleoseismological techniques, among them. These were complemented by the efforts to use archival information to refine locations, rupture extents and magnitudes of large events. We have used much of these data and inferences derived therein, to understand the important seismic sources in the Himalaya and also to review the status of the Central and Assam seismic gaps. Geodetic slip models imply that a great earthquake is imminent in the Central Himalaya, assuming a renewal time of ~500 years and that the 1505 earthquake was the gap-filling penultimate earthquake (Feldl and Bilham, 2006). The large slip of the same vintage (AD 1400) observed by Kumar et al. (2010) is attributed to this event. Based on observations in the trenches excavated in the Nepal Central Himalaya, Lave et al. (2005) reported an AD 1100 event. They assign a magnitude of Mw 8.8 to this event, which caused the coseismic displacement of ~17 (+5/−3) m. Although close to the 1934 source, their trenches did not expose any evidence of younger slip. The trenches, although close to the purported source zone of the 1505 earthquake, did not expose any corresponding slip. The absence of slip from the better-known 1934 event is an indication that ruptures may terminate at the basal detachment–wedge transition. These mid-crustal ramps may remain as geometric asperities during interseismic periods, accommodating convergence through thrusting and folding in the sub Himalaya (Lyon-Caen and Molnar, 1985). Strain accumulation on mid-crustal ramps has been inferred from microseismic and geodetic data from the Central Himalayas of Nepal (Pandey et al., 1995). Earthquakes originating on the splay faults within the wedges may typically be associated with steeply dipping fault planes compared to the gently dipping fault plans of the basal décollement. Thus, the 1991 Uttarkashi and 1999 Chamoli earthquakes, both considered as décollement earthquakes, originated on gently dipping (10°) fault planes. The

2005 Kashmir earthquake on the other hand originated on a steeper (29°) dipping plane at a shallower source and generated 4–7 m slip. The 1999 Mw 7.6 Chi-Chi (Taiwan) earthquake is another well-known example of large coseismic slip (up to 10 m), caused by failure on a ramp structure (Chen et al., 2001). It has been proposed that as the thrust sheet associated with the rupture bends upward along a ramp, the movement may produce a large vertical displacement leading to the amplified slip (Chen et al., 2001). Eventually large earthquakes could break through these asperities and transfer the deformation to the surface, leading to large surface rupturing events (the Chi-Chi-type events). Thus it is possible that the detachment breaking earthquakes produce no surface ruptures, and those originating on mid-crustal ramps lead to relatively larger displacements. Interpretation of paleoslip observations in trenches in terms of size of the past earthquakes should consider the dual nature of sources. Balanced cross sections from the 1905 Kangra earthquake suggest that the region of rupture termination is marked by a series of thrust faults and folds, from a highly deforming duplex zone. The rupture presumably terminated near the mapped location of the Jawalamukhi thrust fault, a zone noted for overpressurization and higher interseismic strain accumulation (Gahalaut and Chander, 1991; Powers et al., 1998). Thus, two explanations seem possible for the lack of surface rupture for the 1905 earthquake. One, that it occurred on the shallow basal detachment surface as suggested by Seeber and Armbruster (1981), but the rupture terminated before it reached the surface (as suggested in the case of the 1934 earthquake). Two, it occurred on a blind thrust and most of the deformation was through folding (Yeats and Lillie, 1991). Based on historical, archeological and geological records we surmise that the past great earthquake in the Central Himalaya occurred ~1000 years ago and that the gap in seismicity in the Central Himalaya is real. The eastern part of the Central Gap is relatively more active, having generated four earthquakes during the last 700 years (1255–1934). The 1833 and 1866 earthquakes form part of this cluster and the great 1934 earthquake followed these events. The occurrence of 1934 earthquake 101 years after the 1833 event is considered as a departure from the seismic gap theory that postulates specific renewal times for large earthquakes (Bilham, 2004; Feldl and Bilham, 2006). While the renewal time of great earthquakes originating on the detachment fault might reflect the strain buildup due to plate convergence, and thus leading to episodic behavior, the shallow out-ofsequence thrust events result from the deformation of the wedge. Thus, it is likely that the 1833 and 1866 earthquakes originated on a blind thrust, whereas the 1934 event was sourced on the detachment surface. The shallow and relatively steeper splay faults of the wedges are geometrically favorable for the generation of large slip, as observed in the 1999 Chi-Chi and 2005 Kashmir earthquakes, discussed earlier. Evidence for the AD 1100 earthquake with slip ~17 m and the lack of slip during the subsequent events in 1505 and 1934 (Lave et al., 2005)

86

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

argues for the existence of such splay faults within the propagating wedge. Large variations in terrace elevations along Kali Gandaki River (Molnar, 1987a) and the steeper fault planes observed for a few earlier earthquakes in the Nepal area (Baranowski et al., 1984) are suggested as other evidence for active wedge deformation. Some researchers observe that the 1505 earthquake has ruptured the Central Gap (Ambraseys and Jackson, 2003; Kumar et al., 2010). And, assuming a renewal time of ~500 years, the Central Himalaya is considered to be ready for the next great earthquake. Our studies in the area, including several AD 7 to 12 century temples and monuments older than AD 1500 in the Central Himalaya and Delhi, did not lead to identification of any significant shaking effects that could be attributed to a great Central Himalaya earthquake during medieval times. Considering the likelihood that the reported destruction to monasteries separated 700 km apart could also be due to site effects and the possibility for amplification of slip on the blind thrusts, we propose that the status of the 1505 earthquake as a great gap-filling event needs further review. Based on the variety of data presented here we infer that the pre-1803 large earthquake in the Central Himalaya must have occurred ~900 years ago, and that the gap between the 1905 and 1934 ruptures may be real. The 1897 earthquake is associated with a mechanism different from the great décollement earthquakes and its predecessor is probably ~1000 years old. Some of the published age data along with the newly acquired data are presented in Table 1 and are also shown in the space– time diagram (Fig. 10). The ~1000 year-old earthquake in the 1897 source seems to be evident from multiple data; it is one of the few past earthquakes for which archeological and paleoseismological data are available from multiple sources.

Compared to the 1897 event, the paleoseismic record of the 1950 source is sketchy and the available data is inadequate to suggest the time of the previous large earthquake in this region. Paleoseismological investigations in the Lohit valley have exposed evidence for at least two earlier events, one is probably the 1697 Sadiya earthquake and the other one dates to AD 335–781. Although studies in the Assam Valley have led to the identification of sedimentary features generated by some recent events (AD 1400 onwards), evidence for the older events are sketchy. Reddy et al. (2009) reports evidence for an AD 1548 event, but our data is not sufficient to resolve between the features of the 1548 and 1697 events. If the AD 1548 earthquake is similar in size to the 1950 earthquake, future paleoseismological investigations should lead to the discovery of additional secondary effects of the same vintage and perhaps evidence for older events. Should this event be considered as the predecessor of the 1950 earthquake, renewal time of great earthquakes in this region could be much shorter (~500 years). The corridor of low seismicity between the Mikir Hills and the Mishmi Thrust (the Assam Gap) appears to exist because of the occurrence of two large earthquakes associated with these source zones, which are tectonically unconnected. The intervening region, designated as the “Assam gap” is generally devoid of large/moderate earthquakes. We believe that the gap in time and space observed between these tectonically distinct sources is not a seismic gap in the conventional sense. However, in terms of hazard, earthquakes originating from either of these sources pose a great threat to the Assam Valley, as observed from the damage during the 1697, 1897 and 1950 earthquakes. 6. Conclusions 1. The penultimate plate boundary event in the Central Himalaya may have occurred ~1000 years ago. 2. The status of the 1505 event as a great plate boundary event needs review. 3. The temporal gap in the Central Himalaya in terms of recurrence of plate boundary events is real. 4. There is a duality of seismic sources in the Himalaya. The earthquakes originating on the basal detachment plane are not likely to rupture the surface as exemplified by the 1934 Bihar– Nepal earthquake. Those on the out-of-sequence, steeply dipping faults on the crustal ramp may propagate to the surface and lead to large vertical displacements, as observed in the 2005 Kashmir earthquake. 5. Spatial clustering of large earthquakes is observed on the eastern boundary of the Central Gap. The earthquakes originating on the basal detachment may follow episodic behavior corresponding to plate convergence. The out-of-sequence events may be more random as they follow the wedge deformation. 6. The penultimate earthquakes in the 1897 Shillong and 1950 Upper Assam sources may have occurred ~1000 years and ~500 years, respectively.

Acknowledgments

Fig. 10. Space–time diagram showing ages of past earthquakes in the 1897 and 1950 sources detailed in Table 1.

We thank Prof. K.S. Valdiya for introducing us to the rugged terrain of the Central Himalaya. We thank B.P. Duarah, Kakati Bayan, Anil Earnest and Arun Kumar for their help during the field investigations in the northeast India. Vanessa Andrade helped with the figures. This work was initiated through funding from the Department of Science and Technology (DST), Government of India. Current funding for this project is from the Ministry of Earth Sciences, Government of India. CPR thanks the DST for the Ramanujan Fellowship. We thank the two anonymous reviewers whose suggestions have helped to improve this manuscript.

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

References Ambraseys, N.N., Bilham, R., 2000. A note on the Kangra earthquake of April 4, 1905. Curr. Sci. 79, 45–50. Ambraseys, N.N., Douglas, J., 2004. Magnitude calibration of north Indian earthquakes. Geophys. J. Int. 165–206. Ambraseys, N.N., Jackson, D., 2003. A note on the early earthquakes in northern India and Southern Tibet. Curr. Sci. 84 (4), 57–582. Avouac, J.P., Tapponnier, P., 1993. Kinematic model of active deformation in Central Asia. Geophys. Res. Lett. 20 (10), 895–898. Avouac, J.P., Ayoub, F., Leprince, S., Konca, O., Helmberger, D.V., 2006. The 2005, Mw 7.6 Kashmir earthquake: sub-pixel correlation of ASTER images and seismic waveforms analysis. Earth Planet. Sci. Lett. 249, 514–528. Banerjee, P., Bürgmann, R., 2002. Convergence across the northwest Himalaya from GPS measurements. Geophys. Res. Lett. 29 (13), 1652 doi:10.1029/2002GL015184. Banerji, R.D., 1923. Annual report of the Archaeological Society of India, pp. 80–81. Bapat, A., 1985. Comments on “Seismic Gap in Assam, Northeast India” by K. Khattri, M. Wyss, V. K. Gaur, S. N.Saha, and V.K. Bansal. Bull. Seismol. Soc. Am. 75, 1469–1470. Baranowski, J., Armbruster, J., Seeber, L., Molnar, P., 1984. Focal depths and fault plane solutions of earthquakes and active tectonics of the Himalaya. J. Geophys. Res. 89, 6918–6928. Barua, R.K.L., 1966. Early history of Kamarupa. Lawyers's Book Stall, Gauhati, Assam. 234 pp. Ben-Menahem, A., Aboodi, E., Schild, R., 1974. The source of the great Assam earthquake — an interplate wedge motion. Phys. Earth Planet. Inter. 9, 265–289. Bhattacharya, P.M., Pujol, J., Majumdar, B.K., Kayal, J.R., 2005. Relocation of earthquakes in the Northeast Indian region using joint hypocenter determination method. Curr. Sci. 89, 1404–1413. Bilham, R., 1995. Location and magnitude of the 1833 Nepal earthquake and its relation to the rupture zones of contiguous great Himalayan earthquakes. Curr. Sci. 69, 101–127. Bilham, R., 2001. Slow tilt reversal of the Lesser Himalaya between 1862 and 1992 at 78°E, and bounds to the southeast rupture of the 1905 Kangra earthquake. Geophys. J. Int. 144, 1–23. Bilham, R., 2004. Earthquakes in India and the Himalaya: tectonics, geodesy and history. Ann. Geophys. 47 (2–3), 839–858. Bilham, R., 2006. Comment on “Interpreting the style of faulting and paleoseismicity associated with the 1897 Shillong northeast India, earthquake” by C. P. Rajendran et al. Tectonics 25 TC2001. Bilham, R., Ambraseys, N., 2005. Apparent Himalayan slip deficit from the summation of seismic moments for Himalayan earthquakes, 1500–2000. Curr. Sci. 88, 1658–1663. Bilham, R., England, P., 2001. Plateau ‘pop-up’ in the great 1897 Assam earthquake. Nature 410, 806–809. Bilham, R., Gaur, V.K., Molnar, P., 2001. Himalayan seismic hazard. Science 293, 1442–1444. Célérier, J., Harrison, T.M., Webb, A.A.G., Yin, A., 2009. The Kumaun and Garwhal Lesser Himalaya, India. Part 1: structure and stratigraphy. Geol. Soc. Am. Bull. 1262–1280 doi:10.1130/B26344.1. Chander, R., 1988. Interpretation of observed ground level changes due to the 1905 Kangra earthquake, NW Himalaya. Tectonophysics 149, 289–298. Chaudhury, P.D., 1964a. Archaeology in Assam. Department of Archaeology, Gauhati, Assam. 77 pp. Chaudhury, P.D., 1964b. Archaeology in Assam. Department of Archaeology, Gauhati, Assam, p. 77. Chen, W.-P., Molnar, P., 1977. Seismic moments of major earthquakes and the average rate of slip in central Asia. J. Geophys. Res. 82, 2945–2969. Chen, Wen-Shan, Huang, Bor-Shouh, Chen, Yue-Gau, Yuan-Hsi, Lee, Yang, Chao-Nan Yang, Lo, Ching-Hua, Chang, Hui-Cheng, Sung, Quo-Cheng, Huang, Neng-Wei, Lin, Chin-Cheng, Sung, Shih-Hua, Lee, Kun-Jie, 2001. 1999 Chi-Chi earthquake: a case study on the role of thrust–ramp structures for generating earthquakes. Bull. Seismol. Soc. Am. 91, 986–994. Chitrakar, G.R., Pandey, M.R., 1986. Historical earthquakes of Nepal. Bull. Geol. Soc. Nepal 4, 7–8. Cotton, F., Campillo, M., Deschamps, A., Rastogi, B.K., 1996. Rupture F., history and seismotectonics of the 1991 Uttarkashi, Himalaya earthquake. Tectonophysics 258, 35–51. Cunningham, A., 1964. Archaeological survey of India Reports (1862–64): 23 volumes with index, Reprinted by Saujanya Books (1994), 4853pp. Dutta, T.K., 1964. Seismicity of the Assam-zone of tectonic activity. Bull. Natl Geophys. Res. Inst. 2, 152–163. Feldl, N., Bilham, R., 2006. Great Himalayan earthquakes and the Tibetan Plateau. Nature 444, 165–170 doi:10.1038. Gahalaut, V.K., Chander, R., 1991. On the active tectonics of the Dehra Dun region from observations of ground elevation changes. J. Geol. Soc. India 39, 61–68. Gait, E., 1905. A History of Assam. Thacker Spink & Co. Pvt. Ltd, Calcutta. (1963; 3 rd revised edition), 449 pp. Gansser, A., 1964. Geology of the Himalayas. Interscience, New York. 289 pp. Gowd, T.N., Srirama Rao, S.V., Chary, K.B., 1998. Seismotectonics of northeastern India. Curr. Sci. 74, 75–80. Gutenberg, B., Richter, C., 1954. Seismicity of the Earth and Associated Phenomena, 2nd edn. Princeton Univ. Press. 310 pp. Hannay, S.T., 1851. Brief notice of the Sil-sako or stone bridge in Zillah Kamrup. J. Asiatic Soc. Bengal 20, 291–294. Hodges, K.V., 2000. Tectonics of the Himalaya and southern Tibet from two perspectives. GSA Bull. 112, 324–350.

87

Hough, S.E., Bilham, R., Ambraseys, N., Feldl, N., 2005. Revisiting the 1897 Shillong and Kangra earthquakes in northern India: site response, Moho reflection and a triggered earthquake. Curr. Sci. 2 (88), 1632–1638. Iyengar, R.N., Sharma, D., Siddiqui, J.M., 1999. Earthquake history of India and in medieval times. Indian J. Hist. Sci. 34 (3), 181–237. Jackson, M., Bilham, R., 1994. Constraints on Himalayan deformation inferred from vertical velocity fields in Nepal and Tibet. J. Geophys. Res. 99, 13897–13912. Kayal, J.R., De, R., 1991. Microseismicity and tectonics of Northeast India. Bull. Seismol. Soc. Am. 81 (1), 131–138. Kayal, J.R., Sagina, Ram, Singh, O.P., Chakraborty, P.K., Karunakar, G., 2003. Aftershocks of the March 1999 Chamoli Earthquake and Seismotectonic Structure of the Garhwal Himalaya. Bull. Seismol. Soc. Am. 93, 109–117. Khattri, K.N., 1987. Great earthquakes, seismicity gaps and potential for earthquakes along the Himalayan plate boundary. Tectonophysics 38, 79–92. Khattri, K.N., Tyagi, A.K., 1983. Seismicity patterns in the Himalayan plate boundary and the identification of areas of high seismic potential. Tectonophysics 96, 19–29. Khattri, K.N., Wyss, M., 1978. Precursory variation of seismicity rate in the Assam, area, India. Geology 6, 685–688. Khattri, K.N., Rai, K., Jain, A.K., Sinvhal, H., Gaur, V.K., Mithal, R.S., 1978. The Kinnaur earthquake, Himachal Pradesh, India, of 19 January, 1975. Tectonophysics 49, 1–21. Khattri, K.N., Wyss, M., Gaur, V.K., Saha, S.N., Bansak, V.K., 1983. Local seismic activity in the region of the Assa Gap, Northeast India. Bull. Seismol. Soc. Am. 73, 459–469. Kingdon-Ward, F., 1951. Notes on the Assam earthquake. Nature 167, 130–131. Kumar, S., Wesnousky, S.G., Rockwell, T.K., Ragona, D., Thakur, V.C., Seitz, G.G., 2001. Earthquake Recurrence and Rupture Dynamics of Himalayan Frontal Thrust, India. Science 294, 2328–2331. Kumar, S., Wesnousky, S.G., Rockwell, T.K., Briggs, R.W., Thakur, V.C., Jayangondaperumal, R., 2006. Paleoseismic evidence of great surface-rupture earthquakes along the Indian Himalaya. J. Geophys. Res. doi:10.1029/ 2004JB003309. Kumar, S., Wesnousky, S.G., Jayangondaperumal, R., Nakata, T., Kumahara, Y., Singh, V., 2010. Paleoseismological evidence of surface faulting along the northeastern Himalayan front, India: Timing, size, and spatial extent of great earthquakes, 115, p. B12422 doi:10.1029/2009JB006789. Lave, J., Avouac, J.P., 2000. Active folding of fluvial terraces across the Siwaliks Hills, Himalayas of central Nepal. J. Geophys. Res. 105 (B3), 5735–5770. Lave, J., Yule, D., Sapkota, S., Basant, K., Madden, C., Attal, M., Pandey, R., 2005. Evidence for a great Medieval earthquake (~1000 A.F.) in the central Himalayas, Nepal. Science 307, 1302–1305. Lowe, D.R., 1975. Water escape structures in coarse-grained sediments. Sedimentology 22, 157–204. Lyon-Caen, H., Molnar, P., 1985. Gravity anomalies, flexure of the Indian plate and the structure, support and evolution of the Himalaya and Ganga Basin. Tectonics 4, 513–538. Martin, S., Szeliga, W., 2010. A catalog of felt intensity data for 589 earthquakes in India, 1636–2008. Bull. Seismol. Soc. Am. 100 (2), 562–569. Mathur, L.P., 1953. The Assam earthquake of 15 August 1950. In: Rao, M.B. (Ed.), A compilation of papers on the Assam earthquake of August 15, 1950, Geological Survey of India, publication No. 1, pp. 56–60. Middlemiss, C.S., 1910. The Kangra earthquake of 4th April, 1905. Mem. Geol. Surv. India 38, 1–409. Molnar, P., 1987a. Inversion of profiles of uplift rates for the geometry of dip-slip faults at depth, with examples from the Alps and the Himalaya. Ann. Geophys. 5, 663–670. Molnar, P., 1987b. The distribution of intensity associated with the great 1897 Assam earthquake and constraints on the extent of rupture. J. Geol. Soc. India 30 (13–27), 1987. Molnar, P., Tapponier, P., 1975. Cenozoic tectonics of Asia: Effects of a continental Collision. Science 189, 419–426. Molnar, P., Chen, W., Fitch, T., Tapponier, P., Warsi, W., Wu, F., 1977. Structure and Tectonics of the Himalaya: a brief summary of relevant geophysical observations. In: Jest, C. (Ed.), Himalaya: Science de la Terre: Colloq. Int.C.R.S., 268, pp. 269–294. Nakata, T., 1989. Active faults of the Himalaya of India and Nepal. Geol. Soc. Am. Spec. Pap. 232, 243–264. Nakata, T., Otsuki, K., Khan, S.H., 1990. Active faults, stress filed and plate motion along the Indo-Eurasian plate boundary. Tectonophysics 181, 83–95. Nandy, D.R., 2001. Geodynamics of Northeastern India and the Adjoining Region. ACB publications, Lake Town, Calcutta, p. 209. Nath, M., Narain, K., Srivastava, J.P., 1967. The Delhi earthquake of 27th August 1960. Rec. Geol. Surv. Ind. 95 (Pt. 2), 367–382. Nayak, G.K., Rao, V.K., Rambabu, H.V., Kayal, J.R., 2008. Pop-up tectonics of the Shillong Plateau in the great 1897 earthquake (Ms 8.7): Insights from the gravity in conjunction with the recent seismological results. Tectonics 27 doi:10.1029/ 2006TC002027 TC1018. Ni, J., Barazangi, M., 1984. Seismotectonics of the Himalayan collision zone: Geometry of the underthrusting Indian plate beneath the Himalaya. J. Geophys. Res. 80, 1142–1163. Oldham, T., 1883. A catalogue of Indian earthquakes from the earliest to the end of 1869. Mem. Geol. Surv. India 19 (3), 1–53. Oldham, R.D., 1899. Report of the great earthquake of 12th June, 1897, Mem. Geol. Surv. India. reprinted: 1981 Geol. Surv, India, Calcutta. 379 pp. Pacheco, J.F., Sykes, L.R., 1992. Seismic moment catalog of large shallow earthquakes, 1900 to 1999. Bull. Seismol. Soc. Am. 82, 1306–1349. Pandey, M.R., Tandukar, R.P., Avouac, J.P., Lavé, J., Massot, J.P., 1995. Interseimic strain accumulation on the Himalayan crustal ramp (Nepal). Geophys. Res. Lett. 22, 751–754. Poddar, M.C., 1950. The Assam earthquake of 15th August 1950. Indian Miner. 4, 167–176.

88

K. Rajendran, C.P. Rajendran / Tectonophysics 504 (2011) 75–88

Powers, P.M., Lillie, R.J., Yeats, R.S., 1998. Structure and shortening of the Kangra and Dehra Dun reentrants, sub-Himalaya, India. GSA Bull. 110, 1010–1027. Quittmeyer, R.C., Jacob, K.H., 1979. Historical and modern seismicity of Pakistan, Afghanistan, Northwestern India and southeastern Iran. Bull. Seismol. Soc. Am. 69, 773–823. Rajendran, K., Rajendran, C.P., 2000. Earthquake history at selected sites in the Himalaya: recent initiatives, proceedings of the Hokudan International Symposium and School on active faulting, Japan, Proceedings, pp. 393–396. Rajendran, C.P., Rajendran, K., 2005. The status of central seismic gap: A perspective based on the spatial and temporal aspects of the large Himalayan earthquakes. Tectonophysics 395, 19–39. Rajendran, K., Rajendran, C.P., Jain, S.K., Murty, C.V.R., Arlekar, N., 2000. The Chamoli earthquake, Garhwal Himalaya: Field observations and implications for seismic hazard. Curr. Sci. 78, 45–51. Rajendran, C.P., Rajendran, K., Duarah, B.P., Baruah, S., Earnest, A., 2004. Interpreting the style of faulting and paleoseismicity associated with the 1897 Shillong, northeast India, earthquake: Implications for regional tectonism. Tectonics 23 doi:10.1029/ 2003TC001605. Rajendran, C.P., Rajendran, K., Duarah, B.P., Baruah, S., Earnest, A., 2006a. Reply to comment by R. Bilham on “Interpreting the style of faulting and paleoseismicity associated with the 1897 Shillong, northeast India, earthquake: Implications for regional tectonism”. Tectonics 25 doi:10.1029/2005TC001902 TC2002. Rajendran, C.P., Rajendran, K., Duarah, B.P., 2006b. Reply to comment by B.S. Sukhija et al., on Interpreting the style of faulting and paleoseismicity associated with the 1897 Shillong, northeast India, earthquake: Implications for regional tectonism”. Tectonics 25 doi:10.1029/2005TC001942. Ramesh, N.R., Gadagkar, N.S., 1990. Effect of 1950 earthquake on Geomorphology in a part of Brahmaputra valley near Teju, Arunachal Pradesh. J. Geol. Soc. India 35, 87–90. Rao, M.B., 1953. A compilation of papers on the Assam earthquake of August 15, 1950. Geological Survey of India. publication No. 1, 111 pp. Raper, F.V., 1810. Narratives of a survey for the purpose of discovering the resources of the Ganges. Asiatic Res. 11, 446–563. Rastogi, B.K., Chadha, R.K., Rajagopalan, G., 1993. Palaeoseismicity studies in Meghalaya. Curr. Sci. 64, 933–935. Reddy, D.V., Nagabhushanam, P., Kumar, Devender, Sukhija, B.S., Thomas, P.J., Pandey, A.K., Sahoo, R.N., Ravi Prasad, G.V., Datta, K., 2009. The great 1950 Assam Earthquake revisited: Field evidences of liquefaction and search for paleoseismic events. Tectonophysics 474, 463–472. Richter, C.F., 1958. Elementary Seismology. W.H. Freeman Company. Reprinted: S. Chand and Company, New Delhi, 768 pp. Satyabala, S.P., Gupta, H.K., 1996. Is the quiescence of major earthquakes (M ≥ 7.5) since 1952 in the Himalaya and Northeast India real? Bull. Seismol. Soc. Am. 86, 1983–1986. Schelling, D., Arita, K., 1991. Thrust tectonics, crustal shortening and the structure of the Far-Eastern Nepal Himalaya. Tectonics 10, 851–862. Seeber, L., Armbruster, J., 1981. Great detachment earthquakes along the Himalayan arc and long-term forecasting. In: Simpson, D.W., Richards, P.G. (Eds.), Earthquake Prediction-An International Review: Maurice Ewing Series, Am. Geophys. Union, 4, pp. 259–277. Sharma, V.D., 2001. Delhi and its Neighbourhood. Archaeological Survey of India, New Delhi. 161 pp.

Singh, S., Sinha, P., Jain, A.K., Singh, V.N., Srivastava, L.S., 1975. Preliminary report on the January 19, 1975 Kinnaur earthquake in Himachal Pradesh. Earthquake Eng. Study 75 (4) 32 pp. Srivastava, P., Mitra, G., 1994. Thrust geometries and deep structure of the outer and lesser Himalaya, Kumaon and Garhwal (India): Implications for evolution of the Himalayan fold-and –thrust-belt. Tectonics 13, 89–109. Srivastava, H.N., Verma, M., Bansal, B.K., 2010. Seismological constraints for the 1905 Kangra earthquake and associated hazard in northwest India. Curr. Sci. 99, 1549–1559. Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B., McCormac, F.G., v.d. Plicht, J., Spurk, M., 1998. INTCAL98 radiocarbon age calibration 24,000–0 cal BP. Radiocarbon 40, 1041–1083. Sukhija, B.S., Rao, M.N., Reddy, D.V., Nagabhushanam, P., Hussain, S., Chadha, R.K., Gupta, H.K., 1999a. Timing and return period of major paleoseismic events in the Shillong Plateau, India. Tectonophysics 208, 53–65. Sukhija, B.S., Rao, M.N., Reddy, D.V., Nagabhushanam, P., Hussain, S., Chadha, R.K., Gupta, H.K., 1999b. Paleoliquefaction evidence and periodicity of large prehistoric earthquakes in Shillong Plateau, India. Earth Planet. Sci. Lett. 167, 269–282. Sukhija, B.S., Rao, M.N., Reddy, D.V., Nagabhushanam, P., Kumar, D., Lakshmi, B.V., Sharma, P., 2002. Palaoliquefaction evidence of prehistoric large/great earthquakes in North Bihar, India. Curr. Sci. 83, 1019–1025. Sukhija, B.S., Kumar, D.V., Devender, Nagabhushanam, P., 2006. Comment on “Interpreting the style of faulting and paleoseismicity associated with the 1897 Shillong, northeast India, earthquake: Implications for regional tectonism” by C. P. Rajendran et al. Tectonics 25 doi:10.1029/2005TC001852 TC2009. Szeliga, W., Hough, S.E., Martin, S., Bilham, R., 2010. Intensity, magnitude, location, and attenuation in India for felt earthquakes since 1762. Bull. Seismol. Soc. Am. 100 (2), 570–584. Tandon, A.N., 1954. Study of the great Assam earthquake of August 1950 and its aftershocks, Indian. J. Meteorol. Geophys. 5, 95–137. Thingbaijam, K.K.S., Nath, S.K., Yadav, A., Raj, A., Walling, M.Y., Mohanty, W.K., 2008. Recent seismicity in Northeast India and its adjoining region. J. Seismology 12, 107–123. Valdiya, K.S., 1992. The Main Boundary Thrust zone of the Himalaya, India. Ann. Tectonicae 54–84 Supplement to volume 6. Wallace, K., Bilham, R., Blume, F., Gaur, V.K., Gahalaut, V., 2005. Surface deformation in the region of the 1905 Kangra Mw = 7.8 earthquake in the period 1846–2001. Geophys. Res. Lett. 32 (15), L15307 doi:10.1029/2005GL022906. Wesnousky, S.G., Kumar, S., Mohindra, R., Thakur, V.C., 1999. Uplift and convergence along the Himalayan Frontal Thrust of India. Tectonics 18 (6), 967–976. Yadav, R.B.S., Bormann, P., Rastogi, B.K., Das, M.C., Chopra, S., 2009. A homogenous and complete earthquake catalog for Northeast India and adjoining region. Seism. Res. Lett. 80, 609–627. Yeats, R.S., Lillie, R.J., 1991. Contemporary tectonics of the Himalayan frontal fault system: folds, blind thrusts and the 1905 Kangra earthquake. J. Struct. Geol. 13, 215–225. Yeats, R.S., Thakur, V.C., 1998. Reassessment of earthquake hazard based on a faultbend fold model of the Himalayan plateboundary fault. Curr. Sci. 74, 230–233. Zhao, W., Nelson, K.D., Project INDEPTH Team, 1993. Deep seismic reflection evidence for continental underthrusting beneath southern Tibet. Nature 366, 557–559.