Tectonophysics 592 (2013) 130–140

Contents lists available at SciVerse ScienceDirect

Tectonophysics

j ourna l homepage: www.e lsev ie r .com/ locate / tecto

Seismic landscape from Sarpang re-entrant, Bhutan Himalaya foredeep, Assam,India: Constraints from geomorphology and geology

Sujit Dasgupta a, Kiron Mazumdar b, L.H. Moirangcha c, Tanay Dutta Gupta d, Basab Mukhopadhyay d,⁎a Geological Survey of India (GSI), Indiab Geological Survey of India, Chandigarh 160 033, Indiac Geological Survey of India, Shillong 793 003, Indiad Geological Survey of India, 27 Jawaharlal Nehru Road, Kolkata 700 016, India

⁎ Corresponding author. Tel.: +91 33 22520742; fax:E-mail addresses: sujitdasgupta@yahoo.com (S. Dasg

basabmukhopadhyay@yahoo.com (B. Mukhopadhyay).

0040-1951/$ – see front matter © 2013 Elsevier B.V. Allhttp://dx.doi.org/10.1016/j.tecto.2013.02.021

a b s t r a c t

a r t i c l e i n f o

Article history:Received 6 March 2012Received in revised form 2 February 2013Accepted 7 February 2013Available online 20 February 2013

Keywords:Bhutan HimalayaSarpangUltapaniBackthrustActive faultForedeep

Geomorphic landscape and late Quaternary geological attributes from the Raidak–Manas interfluve in theBhutan–Himalayan foothills, Kokrajhar District, Assam led towards documenting the east–west trending,south dipping, 30 km long active Frontal Back Thrust (FBT), well within the foredeep south of the Main Fron-tal Thrust (MFT). Spectacular north facing 6–50 m high tectonic-scarp generated by the north-propagatingemerging thrust front along with a complementary subdued south-facing scarp defines the terrain as apop-up structure. The entire belt is made up of 5 to 8 km wide six distinct blocks, separated by antecedentrivers/streams. Scarp parallel east–west drainage along with linear lakes characterises the emerging thrustfront. Field evidence for a major fault-propagation fold structure along with thrust faulting within thelate-Quaternary fluvial sediments is ubiquitous. Clay beds deposited in lakes along the footwall of FBT haveformed due to blockade of south flowing rivers by episodic upliftment of the hanging wall block; three suchepisodes of uplift since 16 k years correspond to three morphogenic earthquakes of magnitude ~6.9 rupturingthe FBT during late Pleistocene–Holocene. In light of geomorphological and geological studies, neotectonic activ-ity has been modelled as an active south dipping backthrust that originates at shallow crustal depth from southvergent basal Himalayan Decollement in response to the advancing Himalayan wedge.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

The scenario has not changed much so far as logistic facilities areconcerned for studying the geology of eastern Himalaya beyondSikkim, particularly the foothills region since Gansser (1964) pub-lished his monumental work on the Geology of the Himalayas. In con-trast to western and central Himalaya where ‘easier access, betterexposure, climate and less political restrictions’ are conducive to geo-logical investigation, ‘these conditions change drastically from Sikkimto the east with difficult field conditions, more intense monsoon withover 10 m of rain in Assam, restrictions for foreign investigators andborder disputes’ (Gansser, 1993), etc. In spite of these difficulties,Geological Survey of India (GSI) completed systematic mapping ofBhutan (in 1:50 k scale) that resulted some comprehensive publica-tions including the compilation volume edited by Bhargava (1995). Inall such earlier studies including the most recent one by Long et al.(2011), explorationwas focused on the geology of the lesser and greaterHimalaya with cursory reference to the foothill structures, whichevidently bears the signatures of Quaternary deformation, key to

+91 33 22861625.upta),

rights reserved.

downstream analysis on neotectonics, active faults and seismic hazard.It so happens that entire Indo-Bhutan border traverses criss-cross thefoothill region and could neither be studied effectively either from theBhutan or from the Indian side. This is the reality in spite of the factthat Bhutan outer Himalaya is one of the identified seismic pronezones (Bilham et al., 2001). Dense tropical forest, ethnic disturbancesand lack of communication network have compounded the problemresulting in paucity of geological information from the most part of theBhutan Sub-Himalaya. This is in contrast to a wealth of publicationson the Himalayan foothill structure fromwestern and central Himalayaforedeep including Nepal (see among others; Kumar et al., 2006; Maliket al., 2010; Powers et al., 1998; Yeats and Thakur, 2008).

Nakata (1972, 1989) in his pioneering contribution on the geo-morphic history and crustal movements along the Himalayan foothills,studied two re-entrant segments from the eastern Himalaya; the west-ern one, the Jaldhaka re-entrant between Chel-Gish River in the westand Daina River in the east, bordering West Bengal–Bhutan foothillsand the eastern one, Hatisara (hereafter referred as Sarpang) re-entrantalong the Assam–Bhutan border (Fig. 1). In the former, except a narrowpatch of Siwalik sandstone in the mountainous reaches of the JaldhakaRiver, the geology of the re-entrant is dominated by post-Siwalik fluvialfan deposits dissected by a number of east–west tectonic fault scarps,origin of which evidently dates between 33 k and 22 k Ybp (Guha et

Fig. 1. Seismicity Map of Sikkim–Bhutan Himalaya and its foredeep; data source cited in text. Other features are from Dasgupta et al. (2000). Two recent damaging earthquakes are 1September 2009, Bhutan and 2 September 2011, Sikkim. The red rectangle at the centre is the study area of Sarpang re-entrant.

131S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

al., 2007) and possibly represents different strands of the Main orHimalayan Frontal Thrust (MFT or HFT). Some of the important faultscarps mapped by Nakata (1972, 1989) include the Matiali and Chalsasouth facing scarps while Bharadighi and Thaljhora north facing scarps.The latter scarp between the Jiti River and Bhutan border is a gigantic60 m+ high cliff above the Jiti River bed, 6 km wide (within India)from west to east and can be traced for about 8 km from the scarptip to Nagrakata in the south before it merges with younger alluvium.A notable feature in the vicinity of the north-facing scarps is the pres-ence of lake–marshy land immediately north of the scarp indicatingblocking of the south bound Himalayan Rivers, which could only resultfrom sudden and episodic uplift of the block to dam the river.

The geomorphology of the western part of the Sarpang re-entrant(red box in Fig. 1) was also studied by Nakata (1972, refer to Fig. 15 inNakata, 1989). He named the different geomorphic blocks from westto east as Ripu–Singimajli–Lalbheti–Saralbhanga and Ultapani blocks, allcharacterised by and terminated to the north bynorth-facing scarpswithheight decreasing from west to east and the presence of marshy land–lake to the immediate north of the scarp. Nakata (1972) further mappeddifferent geomorphic surfaces which were classified as 1) Alluvial Plain,2) Saralbhanga River Surface, 3) Singimajli Surface and 4) Lalbheti Sur-face primarily based on the colour profile developed on the surfaces. Hedescribed the scarps as tectonic but stopped short of drawing inferenceson the nature and origin of the structures in the overall framework of theOuter Himalayas, to the foreland of the Main Boundary Thrust (MBT).

In the present paper, we intend to summarise the geomorphologicaland geological work carried out by GSI intermittently between 2006and 2010 within the Indian part of the ‘Sarpang’ re-entrant to highlightactive fault features (that is, the seismic landscape of the area;e.g., see Michetti et al., 2005) in the area in terms of seismic hazards.Our primary goal is to document the morphotectonic features fromthe area and interpret the major structure as an active ‘backthrust’(dipping towards foreland), which is not commonly documentedfrom the Himalayan foredeep.

2. Seismicity in the Region

Earthquake activity from the region (26°–28° N/ 88°–92° E) wascompiled from Reviewed ISC Bulletin for the period 1900–2010; thecatalogue was updated to include events till March 2012 from NEIC

database. The reviewed ISC bulletin also includes data from local net-work (RRLJ), thus threshold magnitude from themid nineties is around3.0 mb. A total of 260 earthquakes are plotted (Fig. 1) on amap showingmajor faults, adopted from the Seismotectonic Atlas of India (Dasguptaet al., 2000). Concentrations of moderatemagnitude earthquakes locatein the Higher Himalaya domain both in the northwest (Sikkim) andnortheast (eastern Bhutan). In the former area, themost recent damag-ing earthquake is that of 18th September 2011, M 6.9 that killed morethan 100 lives; in eastern Bhutan the earthquake of 21st September2009, M 6.0 was also damaging. The foredeep region is characterisedby the occurrence of dispersed smaller magnitude earthquakes; thoughnot very conspicuous, a concentration of events is located south of thepresent study area around the inselbergs (Fig. 3).

3. Geomorphology and geology of Sarpang re-entrant

The spectacular geomorphic and structural features within thewestern portion of the Sarpang re-entrant between Pinkhua Kholain the west and Leu Pani in the east (Figs. 1 and 5) were revisitedfor the first time after Nakata (1972) who worked in the area in themid nineteen sixties.While Nakata did not have access to Bhutanduringworking from India, Gansser whowas working in Bhutan in the thirtiesand after, possibly did not have direct entry facility across the border.In the area, Siwalik rocks are exposed in the salient up to the westernfringe of the Sarpang re-entrant across the Indian border and then dis-appear below the alluvium cover of the Saralbhanga River along withthe northerly retreat of the mountain front. That the central BhutanSub-Himalaya structures were out of sight of Gansser is clearly broughtout from his sketch (Fig. 2, Gansser, 1964, p 194); evidently hidden bythe lesser Himalaya ranges when viewed from Bhutan. As can be seenfrom the DEM, the central Bhutan foredeep is quite narrow, around35 km between the Indian shield Precambrian inselbergs and the foot-hills (Fig. 3). The raised but dissected alluvial blocks extending fromwest (from west-central part of the image frame) to east for about30 km and around 5 to 8 km across clearly show the advancing fore-land deformation front in this part of the Outer Himalaya. Maximumelevation of the easternmost Ultapani structure is around 200 m(above msl), while it is of the order of 240–260 m in other westernblocks; this is in contrast to 60 m elevation of the alluvial plain infront of the inselbergs. South of the inselbergs, bank elevation of

Fig. 2. Sketch of Bhutan foothills (after Gansser, 1964). Note that the Sarpang structures discussed are out of sight when viewed from north and are hidden between ‘SW’ and ‘W’

in the sketch.

132 S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

the Brahmaputra River is 30 m (above msl). The present elevation ofthe uplift terrain is achieved in possibly three phases of vertical riseas deciphered from break in slopes in east–west topographic profile(Fig. 4) along the structures.

A schematic geomorphic-cum-geological map of the area (Fig. 5)shows that the uplifted terrain branches out from the convex endof the Siwalik rocks, west of Pinkhua Khola. The northern boundaryof elevated terrain is marked by sharply defined north facing scarpvarying in height between 55 m in the west (Pinkhua block) and25 m in the east (Ultapani block). The terrain gently slopes to thesouth to be truncated by another south facing smaller scarp of 5–15 mhigh or by a sharp gradient to the south. Thus the entire Pinkhua–Ultapani 30 km long elevated platform is bounded by two oppositefacing scarps. The uplift terrain is separated into a number of blocksby south flowing antecedent rivers. West of Hel River, the Pinkhuaand Ripu blocks are much dissected and degraded with loss of originalscarp height and shape; the Singimajli block is undergoing dissection

Fig. 3. DEM of the Sarpang re-entrant. The elevated sub-Himalayan landforms are visible nejust 30 km south of the elevated structures.

with the development of N–S erosional scarp and 1st order drainage,and in fact a major portion of the south-western segment of the blockhad already been eroded away by the Hel River. Compared to its west-ern part, the Lalbheti block is well preserved and the Ultapani dome ap-pears to be a nascent structure. The overall geomorphic features andstages of terrain evolution pattern along and across the elevated struc-ture prompt us to conjecture that the structure is propagating to theeast and partly concealed below the alluvium beyond Leu Pani.

The most interesting geomorphic feature in the area is the de-pressed region located all along at the base of the north-facing scarp(Figs. 6a and7a) that is presently dotted by anumber of lakes andmarshyland. More than a metre thick lacustrine clay deposit characterises thebase of the scarp, in contrast to very coarse clastics that constitute thescarp. In fact more than one layer of such clay deposit is noticed withinthe subsiding block, north of the Lalbheti scarp. A few east–west linearlakes are also located along the southern endof the elevated structure inthe Ripu–Pinkhua segment.

ar the west-central part of the frame. Note the Indian shield rock inselbergs are located

Fig. 4. E–W topographic profile (VE 45×) along the elevated structure under study; for location of the profile, see Fig. 5 inset. Six elevated blocks from Pinkhua in the westto Ultapani in the east are marked. Material constituting the uplift is late Quaternary unconsolidated fluvial fan deposit. Note: break in slopes within individual blocks indicatesepisodic rise of the terrain. For plan view, see Fig. 5.

133S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

Salient geomorphic and structural features from each of the upliftedblocks are discussed separately as the entire domain has not beenuplifted uniformly in a single episode; different blocks have sufferedvaried amounts of crustal movement and subsequent erosion, resultingin unique morphology to each block (see Figs. 4 and 5).

3.1. Ultapani block

The Ultapani block (Fig. 6a) represents an oval to elliptical shapeduplift-structure between south flowing Samukha Nadi in the west andLeu Pani in the east. The east–west long axis of the uplift block is of theorder of 6.5 km while its north-south extent is around 4.5 km. Promi-nent scarp with varying heights is developed along themargin to definethe elevated ridge. Materials constituting the Ultapani landform include

Fig. 5. Geomorphology-cum-geological map of Central Bhutan Himalaya and foredeep. In tyellow colours respectively. The white portion covers recent fluvial sediments including pblock; 3 — Lalbheti block; 4 — Singimajli block; 5 — Ripu block and 6 — Pinkhua block. FroMBT — Main Boundary Thrust; MFT — Main Frontal Thrust; FBT — Frontal Back Thrust and thsection lines (Figs. 4 and 10a, b, c).

floating boulders and cobbles set in an oxidised silty–clay matrix. Thenorthern scarp of the Ultapani block is quite distinctive with heightvarying from 10 to 25+m (Fig. 6c) above the level of Tarpara Jhora, astream that originates from the marshy lake located at the base of thenorth-facing scarp. Streams flowing from the marshy land travels bothto thewest and east to join the SamukhaNadi and Leu Pani respectively.This feature of a stream apparently flowing in opposite directions givesthe name Ulta (opposite) pani (water), a village with this name locatedwithin the reclaimedmarshy land. The original lake spanning the entirebase of domal uplift is now restricted to the northwestern corner of thesag, locally known as Mach Bhandar (abode of fish; Fig. 6b). Along withthe south facing 5 to 8 mhigh southern scarp (Fig. 6d), theUltapani struc-ture resembles a ‘wrinkle ridge’ in plan view (see Fig. 4 from Schultz,2000) that represents an elevated structure, bounded by two conjugate

he foredeep domain higher and lower elevated landforms are depicted by brown andatch of lacustrine clay occurring just north of: 1 — Ultapani block; 2 — Saralbhangam north to south the important structural elements are: MCT — Main Central Thrust;e southernmost south verging thrust conjugate to FBT. Inset map shows the geological

Fig. 6. Geomorphic and structural features developed in and around the Ultapani block. a) Elevated elliptical Ultapani wrinkle ridge (imagery from Google Earth) and theSaralbhanga block in the west; note marshy land bordering the northern fringe. b) View of Mach Bhandar Lake with lacustrine deposit (youngest of the three such deposit; seetext) along the northwest corner of Ultapani ridge. c) 20 m high Ultapani north-facing fault scarp; view from the southern bank of Tarpara Jhora, east of Ultapani village.d) South facing 8 m high fault scarp; view from north along the Ultapani–Bishmuri Road; note the distant view of road passing through lower elevation. e) Schematic N–S sectionshowing the structural configuration across the Ultapani block.

134 S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

thrusts. N–S profile across the Ultapani ridge (Figs. 6e and 10a) clearlytakes the form of a fault-propagation fold with a steeper northern fore-limb and a gentler long backlimb culminating against a north dippingconjugate thrust. The main south dipping thrust, which has been re-ferred to as the Frontal Back Thrust (FBT) nucleates from the regionaldecollement (Fig. 10a). On the ground the FBT is marked by the east–west stream, debouching on either side just below the north-facingscarp; followed by themarshy land to the north. Lacustrine clay depositof 0.5 to 1.5 m thick overlies coarse sand and boulder bed in the FBTfoothill block. As will be discussed later this is the youngest clay depositof the three different lacustrine sequences identified through differentstratigraphic levels from the footwall domain of other blocks.

The 10 kmwide area between the FBT in the south and the Bhutanhills in the north is covered by younger unconsolidated boulder bed.Siwalik sediments are not exposed in the area and the Daling–BuxaLesser Himalayan rocks directly come in contact with the HoloceneSaralbhanga–Dhol Pani–Bhur River interfluve material, marking the

location of the Main Boundary Thrust (MBT) (Fig. 5). In the absence ofany Siwalik sediments the location of the Main Frontal Thrust (MFT)as depicted (Figs. 5 and 10a) is approximate, only with geomorphicconstraints; the drainage division of the few incipient 1st order streamsthat originate around 200 m contour and flow southerly to join theeast–west stream via the marshy land, possibly defines the surface ex-pression of MFT within the re-entrant (compare location of MBT andMFT by Long et al., 2011). The 6.5 kmwide zone between the northerlydipping MFT and southerly dipping FBT is developing as a sag basindestined to forma structural valley like present dayDehradun inGarhwalHimalaya, India.

3.2. Saralbhanga block

This is a deceptively depressed 2.5 km wide block (Figs. 4–6) be-tween the Ultapani wrinkle ridge in the east and the uplifted Lalbhetiblock, bordering the Saralbhanga River in thewest. N–S extent is around

Fig. 7. Geomorphic and structural features developed in and around the Lalbheti–Singimajli Block. a) Plan view of the uplift blocks (imagery from Google Earth) showing the northfacing scarp; light coloured elongated patches to the immediate north of scarp are remnant lakes with lacustrine sediment. b) and c) Lake sediments on top of the Singimajli andLalbheti footwall blocks respectively (stratigraphically middle of the three lacustrine sequence; see discussion in text). d) Vertical forelimb with stretched and steeply plungingpebble-cobble bed in the hanging wall of the regional fault-propagation fold structure; location shown with black arrow. e) Fresh cut river section shows the thrust contact betweenthe vertical forelimb and sub-horizontal backlimb.

135S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

4.0 kmwith a distinct 6 to 10 mhigh northern scarp against themarshyland (refer Fig. 6a) in the north. There is no visible scarp defining thesouthern end of the block. Apart from the difference in height fromthe adjacent eastern and western uplifted blocks, the top surface ofthe area is free from oxidised soil cap that characterises the upper por-tion of the Ultapani and Lalbheti blocks. As will be discussed later, themajor part of this block was eroded away consequent to a high energyflood that resulted from a pre-historic dam (blocking the SaralbhangaRiver) break following a major earthquake.

3.3. Lalbheti block

This is the most conspicuous elevated structure (Fig. 4) in the areathat has preserved sufficient evidence on the origin of these uplifted

blocks well within the foreland, south of the MFT. Along its central por-tion the block extends for about 5 km from north to south. At its north-ern end, between the rivers Singimajli in the west and Saralbhanga inthe east it is about 3.75 kmwide, while at its southern end its E–W ex-tent is about 5.0 km between the Hel and Saralbhanga Rivers (Fig. 5).The gigantic 35 to 50 m high north facing scarp (Fig. 7a) overlookingthe Bhutan Himalaya is impressive. The top surface of the Lalbhetiblock with oxidised red soil slopes to the south, terminating against a10 to 15 m high south facing scarp. While the north and south facingscarps copiously demonstrate fault scarps, the eastern and western N–Sscarps along the Saralbhangha and Singimajli rivers are developeddue to river cutting by these antecedent streams. At least two slopebreaks (Fig. 4) can be identified from the riverbed level to the top of struc-ture possibly indicating two major movements in attaining the present

136 S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

elevation. To the immediate north of the scarp there are a number ofelongated lakes (Fig. 7a) with thick clay deposit (Fig. 7b, c).

In the northwestern corner of the Lalbheti block, where it is cutand eroded by a new channel of the Singimajli River, steeply dippingbeds of siltstone with layers of cobble-boulder are exposed on theriver bed while sub-horizontal beds of similar lithology exposed at ahigher elevation (Fig. 7e). The contactmarks as a southerly dipping thrust(backthrust) between steeply dipping forelimb and sub-horizontalsegment of the fault-propagation fold. The forelimb at its thrust contactis characterised by stretched and steep long-axis plunge of cobbles-boulders (Fig. 7d). A schematic section (Fig. 9) across the Lalbhetiblock depicts the structure.

The entire north–south section of the Lalbheti block (both theuplifted hanging and subsided footwall segments) is exposed alongthe western bank of the Saralbhanga River (Fig. 8a). Stratified, hori-zontal bedded pebble, cobble and boulder bearing coarse sand consti-tute the 30–35 m exposed river section (Fig. 8e); as one approachesthe northern scarp, the beds dip 10–15° to the north before exposingsteeply dipping and indurated boulders and coarse sediments (Fig. 8f).This change from sub-horizontal to sub-vertical attitude occurs withina distance of 40 m and the fold form-surface can be reconstructed fromorientation of long axes of the boulders with steep plunge and evenoverturned to the south. Stretched cobbles, similar to that describedfrom the northwest side (Fig. 7d), are seen within the extreme rightband in Fig. 8f; the thrust zone dipping to the south passes throughthe base of the exposure with the development of extensive induratedfault gouge (Fig. 8g). The structure at the northeastern corner of theLalbheti block thus suggests tip of a fault-propagation fold with sub-

Fig. 8. Geomorphology, structure and sediment type in the Saralbhanga River section alongthe Lalbheti elevated block showing the uplifted hanging wall (35 m), depressed footwallhanging wall showing stratified coarse sediment. c) and d) Enlarged view of the footwall(note location of sample for age dating in figure). e) Enlarged view from (b) showing the lowf) Enlarged view frommiddle part of the hanging wall at distance of 40 m from (e) showing sof coarser sediment; note contact with pebble-sand-silt layer to the extreme right of the phothe bottom of the photo. g) Indurated fault gouge.

vertical to overturned forelimb and a sub-horizontal backlimb. Thefirst-order topography of the area (Fig. 10b) clearly attests to the re-gional structure (Fig. 9) with south dipping back-thrust (FBT) rupturingthe surface.

Similar to the situation at the base of the Ultapani block, the baseof the northern scarp of Lalbheti block is depressed with east–westflowing streams (Tractor nalameeting the Saralbhanga River) and pres-ence of extensive lake (locally known asMagur Bil, Fig. 7c). North of thenorth-facing Lalbheti scarp, i.e., in the footwall of the FBT, thick darkgrey clay bed is exposed (Fig. 8c and d) at the bottom of 5 m footwallsection along the Saralbhanga River overlain by very coarse-grainedpebbly cross-bedded sand horizon. This 1.5 to 2.0 m thick clay bedamid very coarse fluvial detritus along the Himalayan foothills indi-cate deposition of sediments in lakes formed by damming of southerlyflowing rivers (Saralbhanga in this case) during episodic rise of theblocks through movement along the FBT. From relative stratigraphicconsideration this clay bed is the oldest among the three lacustrinehorizons referred above. Five samples were collected from this claybed (Fig. 8d) and C14 ages (Table 1) analysed. The age of this preservedlacustrine clay being of the order of 16,000 (2 sigma calibrated radiocar-bon age) years indicates the time of the earliest earthquake that upliftedblocks 2 to 6 (see Figs. 4 and 5) to dam the southerly flow of themightyproto-Saralbhanga River. Subsequent breaching did erode the majorportion of the Saralbhanga block and the present 6 to 10 m northernscarp (see above) is the youngest scarp resulting from the latest Holo-cene event. The name ‘Lalbheti’ is suggestive; while Lal means red,bheti in local language means barrier. A barrier (dam) of red colouredearth was recognised by the early settlers in the area.

the eastern side of the Lalbheti Block (for location refers Fig. 7). a) Panoramic view of(5 m) and the emergent thrust front. b) Enlarged view of the northern segment of thesection showing the lower clay deposit (16 ka; see text) overlain by coarser depositer part of the hanging wall with unconsolidated stratified pebble–cobble–sand deposit.teeply dipping to overturned forelimb fold structure defined by orientation of long axesto; the shallow south dipping thrust plane passes through the freshly cleared area near

Fig. 9. Schematic section showing the Lalbheti fault propagation fold structure with thrust rupturing and exposed to the surface at the emergent thrust front.

137S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

About 2.25 km north of the north-facing Lalbheti scarp, UpperSiwalik boulder beds are exposed that overrides the distal end of theFBT footwall (Figs. 5 and 10b). This marks the location of MFT and thedepressed region between the FBT in the south and MFT in the northdefines the ‘dun-type’ valley. It may be noted that similar valley infront of the Ultapani ridge was much wider to the tune of 6.5 km.

3.4. Singimajli and Ripu blocks

At its northern scarp end, east–west extent of the Singimajli blockis around 4.0 km; while it is of the order of 3.0 km from south tonorth (Fig. 5). Geomorphic terrain evolution of the region clearlyindicates that the Lalbheti and Singimajli blocks uplifted as a singleunit that subsequently got separated by the Singimajli River after itbroke the barrier leaving several lakes (Fig. 7a) with clay deposits(Fig. 7b) and stagnant water. The triangular shape of this block hasresulted through erosion and removal of the material by the southeastflowing Hel River. In fact, the Hel River along with its two upstreambranches, the Phisu Nadi and the Longa River also broke apart theSingimajli block in the east from the remnant linear Ripu block in thewest. Due to these geomorphic changes through fluvial action, thesouth bound scarp is totally obliterated; however the northern scarpis quite conspicuous, with height around 20–28 m above the footwallblock. The same clay bed as noted in the footwall of the Lalbheti blockalso occurs here (Fig. 7b). The N–S extent of the linear Ripu block is3.25 km and a maximum E–W width of 0.5 km. At its northern endmaximum height in the Phisu Nadi section is around 25 m; at its south-ern end though there is no preserved scarp; a few small lakes alignedeast–west should define the location of the subsidiary north dippingthrust.

A schematic N-–S section (Fig. 10c) depicts the structural frame-work of the Outer and Sub-Himalayan domain across the Singimajliblock. In this segment the depressed dun type valley between theMFT and FBT has further reduced to around 2.0 km.

Table 1Radiocarbon age dates of charcoal samplesa from clay (location Fig. 8d).

Sampleno.

Laboratoryno.

ConventionalC14 age

Analyticalerror

2 Sigma calibratedage in ka

071-22b 242776 13460 50 16.32071-22a 242775 13640 50 16.62071-21b 242774 13620 50 16.54071-21a 242773 13540 50 16.43071-20a 242772 13580 50 16.49

a Samples collected from site and age data provided by Dr Doug Yule, California StateUniversity, Northridge, USA.

3.5. Pinkhua block

This is the westernmost uplifted block that splays out from the Hi-malayan Main Frontal Thrust (MFT) located just on the Indo-Bhutanpolitical border. It may however be noted that there is no referenceof this segment by Nakata (1972). The main uplifted block extendsfor about 3.5 km from Pinkhua Khola in the west to Balu Khola inthe east; north–south extent is about 2.0 km. The coarse sedimenttype as exposed in the Balu Khola section is almost similar to thoseencountered in the Saralbhanga River section further east. Topo-graphic map and imagery show a north-facing scarp within the Bhu-tan territory, which was out of bound for physical study. Up to thenorthern limit till one could proceed along the narrow gorge of theBalu Khola the sediment pile was more than 50 m, sub-horizontalto gently dipping to the north that turns highly inclined at its north-ern tip, a situation similar to that described from the eastern end ofLalbheti block. Presence of big boulders of typical Siwalik sandstonewithin the Balu Khola bed indicates the presence of Siwalik outcropjust across the Indo-Bhutan border. The Pinkhua block also gentlyslopes to the south without any major south-facing scarp but 2-5 mhigh short segment scarps are present just south of 160 m topographiccontour.

4. Morphogenic earthquakes

Data and information on the geomorphology and geological attri-butes from the Sarpang re-entrant clearly bring out the contemporarydeformation style in the region which is dominated by a frontal southdipping backthrust (FBT). Surface geological evidence shows devel-opment of a regional fault-propagation fold structure with its steeplydipping forelimb supporting the huge north-facing scarp. The thrustis not blind below the surface fold but has ruptured the surface withsignificant translation. Such backthrust though occasionally presentin the Sub-Himalaya (e.g., see Fig. 6 of Powers et al., 1998), our interpre-tationdiffers from that of the Janauri structure in theKangra re-entrant, inthe sense that it originates from the decollement while the north dippingconjugate thrust abuts against the north verging active Pinkhua–Ultapanifrontal backthrust. The dominant deformation mechanism as could besurmised from the available data is a southward propagating wedgethrust with episodic northward translation along the FBT aided by itsproximity to the rigid shield rocks (Goalpara wedge; see Dasgupta et al.,1987) to the immediate south. Considering the height of Lalbheti hangingwall scarp of 45 m above the footwall clay deposit (16 k years) in theSaralbhanga River section (refer Fig. 8d), the vertical uplift rate (co- andinter-seismic) is of the order of 3.0 mm/year with a steady slip rate of6.0 mm/year along the 30° south dipping FBT. These figures howeverdo not convey the entire scenario when one considers that the footwallclay deposits occur at different levels and thus formed during differenttemporal episodes; at the top of the Lalbheti and Singimajli footwall

Fig. 10. N–S geological section with surface profile along (a) Ultapani (b) Lalbheti and (c) Singimajli blocks; for location see inset Fig. 5. The depth of basement decollementconstrained through data referred in text. The Frontal Backthrust (FBT) originates from the basement at depth ~3.0 km. The uplift blocks are pop-up structures defined by the activeFBT and passive north dipping conjugate thrust. The subsiding domain between the FBT and MFT is referred to as ‘Dun valley’ in text. MFT — Main Frontal Thrust; MBT — MainBoundary Thrust.

138 S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

surface (refer Fig. 7b and c), which occurs about 10 m above theSaralbhanga River bed lower clay (16 ka) that there is another 1.0–1.2 m younger (middle sequence of the 3 deposits) clay deposit. Thisyounger clayey footwall domain supports only grass land in contrast todense forest both to the south (hanging wall) and further north in theBhutan foothills. The third and the youngest claywith partly decomposedtree logs occurs in front of the Saralbhanga and Ultapani blocks (referFig. 6a and b). At present though we do not have any age data for thesetwo blanket clay horizons, they are stratigraphically younger than thelowest dated clay deposit of 16 ka. We postulate that the three claybedsweredeposited in lakes formeddue to blockadeof southflowing riv-ers in response to three episodes of uplift since 16 ka; three episodic up-lifts are preserved in blocks 3 to 6 (see Figs. 4 and 5), albeit erosionaldegradation while only the two younger uplifts can be recognised inblock 1 (Ultapani). Dam breach occurred along block 2 (Saralbhanga),after the second episodic rise, thus only the latest uplift of 6–10 m is pre-served. The 1st uplift at 16 ka resulted from surface rupture length of22.5 km (blocks 2 to 6 in Fig. 4) corresponding to earthquake of magni-tude 6.6 to 6.8 (see Wells and Coppersmith, 1994); the two youngerevents accumulated further slip to produce surface rupture of 30 kmthrough earthquakes of about magnitude 6.8–7.0 each; the youngestevent occurred sometimes during the late Holocene.

5. Foreland structure

The width of foreland fold-thrust belt south of MBT within theSarpang re-entrant is around 15 km and the width of the foredeep isonly of the order of 30–35 km between the exposed Indian shield crustand the sub-Himalaya fold-thrust belt (Fig. 3), the narrowest foredeepsegment along the entire Himalaya. The exposed basement plunges(≈10°) below the alluvium to reach a depth of 2.0 to 2.5 km (deep resis-tivity sounding; GSI, pers. com) between Bishmuri and Bansbari; there-after the regional slope of the basement is of the order of ~5° to definedepth of sub-Himalayan decollement (Powers et al, 1998). In the absenceof any other subsurface data we consider the depth to decollement as aworking model to construct three profiles across the Ultapani, Lalbhetiand Singhimajli blocks (Fig. 10). Accordingly the frontal backthrust(FBT) should nucleate from a depth of 3.0 km over the crystallinedecollement. Such backthrustwedging in foreland thrust belts is usuallycaused by rapid burial beneath syntectonic sediment deposit (Adamet al, 2004). These authors reported landward vergent thrust faultingfrom Cascadia convergent margin offshore Washington and postulatedthat back thrusting results fromheterogeneous diffuse strain accumula-tion in segments with increased sediment input. One major concernis that whether moderate to large earthquake can nucleate from such

139S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

shallowdepths independently or sympathetically participate to rupturealong the Main Himalayan Thrust further downdip. The Kashmir earth-quake of 8 October 2005 (Mw 7.6; surface rupture 70 km) is locatedwithin the sub- Himalaya, a domain same as that of the present studyarea. Though the focal depth for the event is of the order of 13–15 km(NEIC, ISC), major slip has been modelled to nucleate on a blindthrust below a flat decollement at ~5 km depth (Bendick et al., 2007).The Christchrust earthquake of 22 February 2011 (Mw 6.3) also nucle-ated at depth of around 5 km; in an intraplate situation the estimatedcentroid depth for the 30 September 1993 Latur earthquake (Mw 6.1)is of the order of 2.6 km (Seeber et al., 1996). Due to its very shallowfoci all these earthquakes were devastating, and so we prognosticatethat the Sarpang backthrust could generate moderate to large earth-quake from such shallow depths. We however do not preclude thescenario where FBT could originate in the basement from depth of~5 km. Our study indicates that the foreland uplift block occurs as apop-up structure defined by the major north vergent backthrust and asubsidiary north dipping conjugate thrust; the major backthrust hasruptured at least thrice, as indicated by three distinct lacustrine claydeposits at different stratigraphic levels starting from 16 ka, formed dur-ing coseismic uplift to dam the southerly flowing rivers. Wemay furtheradd that the FBT has ruptured onlywestern half of the Sarpang re-entrant(see Fig. 3) and as the pop up structure is propagating to the east futureearthquake (s) can generate surface rupture for another 30 km beforereaching the eastern barrier of the salient. Excess syntectonic dumpingof sediments through the re-entrant within proximal foredeep andpush of theGoalpara basementwedge from the south against the south-ward moving piedmont mass could be the immediate cause for thedevelopment of the active FBT that piggy back on frontal thrust.

6. Discussion and conclusion

Based on geomorphic signatures present within the re-entrantdefined by longitudinal and cross profiles along and across the ter-rain; presence of a 30 km long imposing north facing tectonic scarpthough isolated by erosion and formation of new drainage cuttingacross the elevated pop-up structure; ubiquitous presence of elongat-ed lakes with clay deposits at different stratigraphic levels at the baseof the scarp; presence of exposure level folds and thrust, particularlyin the Lalbheti and Ultapani domain, all points towards presence ofa regional fault-propagation fold structure with the steep forelimbsupporting the scarp varying in height from 6 to 50 m. The fault hasruptured the surface and is therefore not a buried fault as is usuallydocumented from other parts of the sub-Himalaya. This fault hasbeen designated as the Frontal Backthrust (FBT). Sediments constitut-ing the exposed hanging wall are very coarse fluvial boulder–cobble–pebble–coarse sand bedded but unconsolidated late Quaternary se-quence and when deformed are unlikely to maintain and preservethe resultant structural form surfaces. Structures that have been doc-umented and described are through fortuitous encounter withindensely forested hostile terrain. We have identified three lacustrineclay sequences deposited over the footwall block but could date onlythe oldest one that deposited around 16 ka. The younger clay horizonscould be of middle and late Holocene age respectively as decipheredfrom colour and texture of clay and size of associated partly decomposed(older) and relatively fresh (younger) tree fragments. These three claysequences are tagged to three episodic uplifts of the terrain that blockedthe south flowing rivers temporarily and deposit the lacustrine sedi-ments. Unlike passive back thrust documented from other segments ofthe sub-Himalaya, we consider the FBT as an active fault, activity alongwhich has generated three earthquakes during the last 16,000 years.

The width of the Himalayan foredeep in general is several hun-dred kilometres in contrast to the narrow foredeep of 30–35 km inthe present study area. Thickness of Tertiary and older sediments overthe shield crust in Bihar–Nepal foredeep below the sub-Himalaya, is ofthe order of 5 km+ (Dasgupta, 1993), for which at the moment we

donot have sufficient data from the Bhutan foredeep. Depth of the base-ment decollement presented for the Sarpang area based on limited datais conservative and if down-to-basement E–W normal faulting is in-voked to the immediate north of the inselbergs, the depth to basementdecollement below the FBT could be deeper than 3 km or could also in-volve the basement. What we intend to convey is that active fault fea-tures are present on the surface that are obviously related to the FBT,which in our interpretation is capable of generating earthquake inde-pendently. FBT is characterised as a northward moving backthrustpiggy backing on the frontal thrust, which originated from the southvergent basal Himalayan Decollement in response to the advancingHimalayan wedge towards south. Given that the FBT structure mergeswith MBT towards west wedge geometry at crustal level is suggested;similar wedge thrust and associated warps from foreland sequencehave been described in a number of regions where such paired oppo-site vergent thrusts exist (Bendick et al., 2007; Ishiyama et al., 2004;Medwedeff, 1992). The region between back-thrust scarp and the MBTfurther north is being developed as an incipient Dun valley in theBhutan–Assam foredeep; development of such Dun valley with similartectonic motif is widely reported from different parts of the Himalayanforedeep (Philip and Virdi, 2007; Philip et al., 2011; Singh and Tandon,2010; Yeats et al., 1992).

Acknowledgements

Authors acknowledge the short but useful visit to the field by Prof.Roger Bilham (University of Colorado, Boulder, USA) and Prof. DougYule (California State University, Northridge, USA) in late 2007. TheC14 data on the age of clay (Table 1) was kindly provided by Prof.Yule. Interpretation arrived in the paper is however solely those of theauthors. Field visit and encouragement provided byDr B. ChattopadhyayandMr. S. Sanyal, GSI were useful. Assistance provided by Ms. SharmilaGhose and Mr. Sanjib Ghosh in drafting some of the figures was helpful.Constructive comments and interest shown by Dr. Mian Liu, Editorand three learned reviewers including Dr. Alessandro Maria Michettihave considerably improved the quality of presentation and scientificcontent of the paper.

References

Adam, J.A., Klaeschen, D., Kukowski, N., Flueh, E., 2004. Upward delamination ofCascadia Basin sediment infill with landward frontal accretion thrusting caused byrapid glacial age material flux. Tectonics 23, TC3009. http://dx.doi.org/10.1029/2002TC001475.

Bendick, R., Bilham, R., Khan, S.F., Khan, A., 2007. Slip on an active wedge thrust from geo-detic observations of the 8 October 2005 Kashmir earthquake. Geology 35, 267–270.

Bhargava, O.N., 1995. The Bhutan Himalaya: a geological account. Geological Survey ofIndia Special Publication 39 (Kolkata).

Bilham, R., Gaur, V.K.,Molnar, P., 2001. Himalayan seismic hazard. Science 293, 1442–1444.Dasgupta, S., 1993. Tectono-geologic framework of the eastern Gangetic foredeep. Geo-

logical Survey of India Special Publication 31, 61–69.Dasgupta, S., Mukhopadhyay, M., Nandy, D.R., 1987. Active tectonic feature in the

central part of Himalaya. Tectonophysics 136, 255–264.Dasgupta, S., et al., 2000. Seismotectonic Atlas of India and its Environs. Geological

Survey of India, Kolkata.Gansser, A., 1964. Geology of the Himalayas.Wiley Interscience, London/New York/Sydney.Gansser, A., 1993. The Himalayas seen from Bhutan. Jb. Geol. B.-A. Band 136, Heft 2,

pp. 335–346.Guha, D., Bardhan, S., Basir, S.R., De, A.K., Sarkar, A., 2007. Imprints of Himalayan thrust

tectonics on the Quaternary piedmont sediments of the Neora–Jaldhaka valley,Darjeeling–Sikkim Sub-Himalayas, India. Journal of Asian Earth Sciences 30, 464–473.

Ishiyama, T., Mueller, K., Togo, M., Okada, A., Takemura, K., 2004. Geomorphology, kine-matic history, and earthquake behavior of the active Kuwana wedge thrust anti-cline, Central Japan. Journal of Geophysical Research 109, B12408.

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 IndianHimalaya. Journal of Geophysical Research 111, B03304. http://dx.doi.org/10.1029/2004JB003309.

Long, S., McQuarrie, N., Tobgay, T., Grujic, D., 2011. Geometry and crustal shorting ofthe Himalayan fold–thrust belt, eastern and central Bhutan. Bulletin of the Geolog-ical Society of America 123, 1427–1447. http://dx.doi.org/10.1130/B30203.1.

Malik, J.N., et al., 2010. Active fault, fault growth and segment linkage along the Janaurianticline (frontal foreland fold), NW Himalaya, India. Tectonophysics 483, 327–343.

140 S. Dasgupta et al. / Tectonophysics 592 (2013) 130–140

Medwedeff, D.A., 1992. Geometry and kinematics of an active, laterally propagatingwedgethrust, Wheeler Ridge, California. In: Mitra, S., Fisher, G.W. (Eds.), Structural Geologyof Fold and Thrust Belts. Johns Hopkins University Press, Baltimore, MD, pp. 3–28.

Michetti, A.M., Audemard, F., Marco, S., 2005. Future trends in paleoseismology: inte-grated study of the seismic landscape as a vital tool in seismic hazard analyses.Tectonophysics 408, 3–21.

Nakata, T., 1972. Geomorphic history and crustal movements of the foothills of theHimalayas. Tohoku University Science Reports: 7th Series (Geography), 22, pp. 39–177.

Nakata, T., 1989. Active faults of the Himalaya of India and Nepal. GSA Special Paper232, 243–264.

Philip, G., Virdi, N.S., 2007. Active faults and neotectonic activity in the Pinjaur Dun,northwestern Frontal Himalaya. Current Science 92, 532–542.

Philip, G., Suresh, N., Bhakuni, S.S., Gupta, V., 2011. Paleoseismic investigation alongNalagarh thrust: evidence of Late Pleistocene earthquake in Pinjaur Dun, North-western sub-Himalaya. Journal of Asian Earth Sciences 40, 1056–1067.

Powers, P.M., Lille, R.J., Yeats, R.S., 1998. Structure and shortening of the Kangra andDehradun reentrants, Sub-Himalaya, India. GSA Bulletin 110, 1010–1027.

Schultz, R.A., 2000. Localization of bedding plane slip and backthrust faults above blindthrust faults: keys to wrinkle ridge structure. Journal of Geophysical Research 105,12035–12052.

Seeber, L., Ekstrom, G., Jain, S.K., Murthy, C.V.R., Chandak, N., Armbruster, J.G., 1996. The1993 Killari earthquake in central India: a new fault inMesozoic Basalt flows? Journalof Geophysical Research 101, 8543–8560.

Singh, V., Tandon, S.K., 2010. Integrated analysis of structures and landforms of anintermontane longitudinal valley (Pinjaur dun) and its associated mountain frontsin the NW Himalaya. Geomorphology 114, 573–589.

Wells, D.L., Coppersmith, K.J., 1994. New empirical relationships among magnitude,rupture length, rupture width, rupture area and surface displacement. Bulletin ofthe Seismological Society of America 84, 974–1002.

Yeats, R.S., Thakur, V.C., 2008. Active faulting south of the Himalayan front: establishinga new plate boundary. Tectonophysics 453, 63–73.

Yeats, R.S., Nakata, T., Farah, A., Fort, M., Mirza, M.A., Pandey, M.R., Stein, R.S., 1992. TheHimalayan frontal fault system. Annale Tectonicae 6, 85–98.