Tectonophysics 580 (2012) 162–177

Contents lists available at SciVerse ScienceDirect

Tectonophysics
journal homepage: www.elsevier.com/locate/tecto

Late Pleistocene and Holocene large magnitude earthquakes along Himalayan Frontal
Thrust in the Central Seismic Gap in NW Himalaya, Kala Amb, India
G. Philip ⁎, S.S. Bhakuni, N. Suresh
Wadia Institute of Himalayan Geology, 33 General Mahadeo Singh Road, Dehra Dun-248 001, India

a r t i c l e i n f o a b s t r a c t

Article history: The Himalayan Frontal Thrust (HFT) forms the southernmost active tectonic mountain front of the Himalaya. To
Received 16 April 2012 understand the ongoing tectonics further, paleoseismological study has been carried out in the vicinity of the HFT
Received in revised form 11 September 2012 system along the Himalayan Front near Kala Amb, India. The trench excavation survey conducted across an ex-
Accepted 13 September 2012
plicit surface exposure of the HFT exhibits two distinct faults considered to be associated with the reactivation of
Available online 23 September 2012
the HFT where the Middle Siwalik rocks (Late Miocene) have repeatedly thrust over the Late Pleistocene and
Keywords:
Holocene sediments. Presence of large-sized coseismically induced sand-injection feature and its disposition rec-
Active faults ognized in the trench also suggest occurrence of large magnitude earthquakes in this region. An uplifted and
Paleoseismology upwarped strath terrace, 3 to 5 m thick alluvium, resting over the 15 m high Middle Siwaliks, abruptly truncated
Trench excavation by the HFT indicates its latest activity. Optically Stimulated Luminescence dating techniques were employed to
Himalayan Frontal Thrust constrain the chronology of events. The long term slip rate of the abandoned terraces due to the activity of the
Seismic hazard HFT is estimated to be 3.4 mm/yr or greater since Late Holocene. The paleoseismological investigations have pro-
NW Himalaya vided unambiguous evidences of at least two large magnitude earthquakes occurred in this region where an
earthquake with 12 m or larger surface displacement and magnitude 7.5 or greater hit this region in the period
between 29.3 ka and 17 ka in the Late Pleistocene and another great earthquake recurred with 20–22 m or more
surface displacement and magnitude of 7.7 or greater between 5.8 ka and 2 ka in the Holocene. The present
study is the first time report of multiple large magnitude paleoearthquakes in the northwestern part of the
Frontal Himalaya during Late Pleistocene and Holocene. The repeated reactivation of HFT substantiates high seis-
mic potential of the Frontal Himalaya and calls for more extensive study of paleoearthquakes of this vastly pop-
ulous mountainous region.
© 2012 Elsevier B.V. All rights reserved.

1. Introduction
The collision between the Indian and Eurasian plates around 50 Ma
along the Indus–Tsangpo Suture Zone (ITSZ) has subsequently resulted
to the uplift of the Himalaya, the highest mountain belt of the world.
The plate motion models and GPS based measurements indicate that
the convergence between the Indian and Eurasian plates is about
40–50 mm/year (Banerjee and Bürgmann, 2002). Of these, ~10 to
20 mm/yr of convergence is accommodated along the entire Himalayan
arc and the remaining is accommodated beyond the Himalaya towards
the north. The continued convergence of the Indian plate has produced
linear zones of deformation, resulting in crustal shortening, especially
along the prominent boundary faults of the Himalayan orogenic belt
(Gansser, 1964; Lyon-Caen and Molnar, 1983; Seeber and Armbuster,
1984; Valdiya, 2003). These crustal scale major tectonic features identi-
fied as the Main Central Thrust (MCT), Main Boundary Thrust (MBT)
and Himalayan Frontal Thrust (HFT), have contributed to the present
day structural and topographic architecture of the fold and thrust belt
⁎ Corresponding author. Tel.: +91 135 2525407; fax: +91 135 2525200.
E-mail address: gphilip@wihg.res.in (G. Philip).
of the Himalaya. All the above NW–SE striking south verging thrusts be-
lieved to be merging into a common decollement, the Main Himalayan
Thrust (MHT) (Brown et al., 1996; Nelson and Project INDEPTH Team,
1998; Seeber and Armbuster, 1984). As a result of ongoing convergence
along the Himalayan Front and its influence in the Outer Himalaya,
many imbricate structures have also been developed subsequently
(Fig. 1). The convergence and shortening have been accommodated
along various thrust planes and have eventually resulted in generation
of numerous seismogenic active faults along the deforming Himalayan
Front.
The historic records of past seismic events in the Himalaya are poor-
ly documented and only a very few site-specific paleoseismological
studies have been carried out to recognize causative active faults and
the paleoearthquakes. For example, in the Outer Himalaya or the Foot
Hills (Fig. 1), lying between the MBT and HFT, so far few active faults
have been reported (Kumar et al., 2001, 2006; Malik and Mathew,
2005; Malik and Nakata, 2003; Malik et al., 2003, 2008, 2010; Nakata,
1972, 1989, Philip and Virdi, 2006; Philip et al., 2011 and references
therein), which have generated major earthquakes. A trench excavation
survey across the HFT near Kala Amb reports evidence of two large
magnitude earthquakes during the past 650 years subsequent to

0040-1951/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.tecto.2012.09.012
 G. Philip et al. / Tectonophysics 580 (2012) 162–177 163

Fig. 1. Regional geological set up and seismicity of the Himalaya. a. Map showing the Himalayan Frontal Thrust (HFT) and the significant earthquakes (Ambraseys and Douglas,
2004; Khattri, 1987; Quittmeyer and Jacob, 1979; Rajendran and Rajendran, 2005). b. Major thrust faults (MHT, MCT, MBT and HFT) and meizoseismal zones (shaded and labeled
with year) of major historical earthquakes along the Himalayan arc and location of the study area. c. Generalized NE–SW geologic section (N–S in Fig. b) across the central portion of
the Himalayan arc (Seeber and Armbruster, 1981).

1294 A.D. and 1423 A.D., and another possible rupture around 260 A.D.
(Kumar et al., 2001). Evidence of two parallel to sub-parallel
active-fault traces along the Himalayan Front around Chandigarh has
provided additional information on the imbricated faulting pattern that
branches out from the HFT system (Malik et al., 2003, 2008). The HFT
near Chandigarh (Fig. 2) has repeatedly ruptured in the past as is evident
from fault scarps with heights varying from 15 to 38 m. The Optically
Stimulated Luminescence (OSL) dates suggest the occurrence of a
major earthquake around 1300–1400 A.D. Malik et al. (2010) have car-
ried out trench excavation survey along the northwestern end of the
Janauri anticline (Fig. 2) in the foothill zone and have reported occur-
rence of a large magnitude earthquake around 400 years ago. Data of
micro-seismicity from temporary/permanent stations has also shown
high percentage of clustering of microearthquakes in some areas of
northwestern Himalaya (Kayal, 2001). However, with limited instru-
mental data and poorly documented historic records of large and moder-
ate earthquakes, it is difficult to reasonably evaluate seismic potential of
any region along the ~2500 km east–west stretch of the Himalayan arc.
As per the available information on seismicity and GPS based
monitoring of crustal movements, the thickly populated areas of the
Ganga plain, Outer and Lesser Himalayan realms are prone to major
earthquakes. The area between 1905 Kangra and 1934 Bihar–Nepal
earthquakes has been categorized as the Central Seismic Gap
(Khattri and Tyagi, 1983) and has a high probability for one or more
M > 8 Himalayan earthquakes in this century (Bilham et al., 2001).
The losses in terms of life and property would be much higher than
for the great earthquakes experienced in the past hundred years be-
cause of the explosive growth of population and industrial establish-
ments in the Frontal-Outer Himalayan realm in the last half a century.
During the last 600 years, along the whole Himalayan arc, four sig-
nificant earthquakes occurred which are either instrumentally or his-
torically documented (Quittmeyer and Jacob, 1979). They are the June
12, 1897 Shillong (Mw~ 8.1); April 4, 1905 Kangra (Mw7.8); January
15, 1934 Bihar–Nepal (M~ 8.3), and August 15, 1950 Assam (M~ 8.7).
The existing historical records also supplement a few more major earth-
quakes in the Himalaya which are estimated to be M> 7.5. These are
the September 1, 1803 Garhwal (M 7.7); August 26, 1833 Kathmandu
(M 7.7); August 28, 1916 western Nepal (M 7.3); July 29, 1947 Assam
earthquakes (Ambraseys and Jackson, 2003).
Evidence of recurring paleoearthquakes in the northwestern Frontal
Himalaya is therefore significantly important for recognizing the
earthquake hazard of these zones. Based on paleoseismological investi-
gations, we present here evidences of large magnitude multiple
paleoearthquakes that have produced surface rupture along the HFT,
in the northwestern Himalayan front, near Kala Amb, Himachal
Pradesh, India (Figs. 2 and 3). The earthquakes to occur within a future
time span of concern to society will be therefore very important in the
highly populous region of the Frontal and Foot Hills of the northwestern
Himalaya.
1.1. Data sources and methodology
CORONA satellite photographs taken October 1965, multi-spectral
satellite data of IRS-ID-LISS-III and PAN acquired 04 October 2002,
CARTOSAT-1 acquired 08 June, 2007 and black and white aerial pho-
tographs taken 1988 of the Frontal area constituted the main data
source for the present study. The DEM for part of the study area
taken from the SRTM data has been useful to map some of the
164 G. Philip et al. / Tectonophysics 580 (2012) 162–177

Fig. 2. Regional location map (SRTM image) showing major duns in the Outer Himalaya and the study area near Kala Amb. Inset: regional geological set up of part of the Outer
Himalaya between the rivers Ganga and Satluj.

major structures. Additionally, the satellite images provided by was carried out across the HFT at Singhauli village, near Kala Amb. To-
Google™ have also been found reasonably useful as they could pro- pographic profiles across the fault have been prepared, using Total
vide the 3D expressions of the terrain. A trench excavation survey Station, to understand the variation in the ground surface levels due

Fig. 3. Geological map: a. Geological map of the area showing major lithological and structural features of the Frontal Himalaya around Kala Amb. The Kala Amb Fault has dextrally displaced
the Siwaliks (Srivastava et al., 1981). b. Schematic cross section across the HFT in the study area, Kala Amb showing the Siwaliks thrusting over Indo–Ganga Alluvium (IGA). c. Corona sat-
ellite photograph showing the Kala Amb Fault and the trench location near Singhauli in the Frontal Himalaya. Location of the strath terrace of the Markanda River is also shown in the figure.
 G. Philip et al. / Tectonophysics 580 (2012) 162–177
to the multiple fault activities. Selected samples of the sedimentary
units collected from the trench as well as from the fluvial terraces
were dated by OSL technique at the Wadia Institute of Himalayan
Geology, Dehradun, India.
1.1.1. Geological setting
The study area is situated in the northwestern Frontal Himalaya at
Singhauli village, ~10 km southeast of the Kala Amb township (Fig. 2).
The HFT in the Kala Amb area is a 20–30°N or NNE dipping thrust. It
brings the Tertiary rocks (here the Middle Siwalik sandstones) over
the Quaternary alluvium (here the Indo–Ganga Alluvium) in the pied-
mont zone (Fig. 3). Active deformation along the HFT is recognized by
the presence of fault scarps (although presently scarps are degraded
and modified to some extent), and uplifted and back tilted Late Pleisto-
cene and Holocene deposits. Ongoing land modification in the region
has suppressed the tectonic landform features and presently very subtle
and discontinuous expression of the fault is identifiable on the surface.
The HFT, in general, trends NW–SE and has also been displaced by
several transverse faults. The Kala Amb transverse fault near Kala
Amb, has apparently displaced the surface trace of the HFT by over
5 km (Fig. 3). Geologically and topographically the area can be divided
into two broad segments separated by the HFT viz. the Siwalik foreland
terrain to the north and the piedmont zone to the south (Fig. 3).
The Siwalik sediments of the Neogene foreland basin in the study
area are represented by the gray and brown sandstones and mud-
stones with bedding-parallel layers of cemented concretionary layers
that are characteristic of the Middle Siwaliks. The rocks are folded into
NW–SE trending open folds which developed on the limbs of the re-
gional Dhanaura anticline (Fig. 3a) with its hinge line passing about
2 km north of Singhauli village. About 3 km east from the present
study area, the Dhanaura anticline has been described as an open cy-
lindrical fold with limbs dipping 20–25° towards NNE and SSW direc-
tions (Raiverman et al., 1983; Srivastava et al., 1981). However, in the
study area, this fold has become south verging overturned asymmetric
fold due to displacement along the HFT (Fig. 3b). The Siwaliks, ex-
posed in many deep cut streams, display complex deformation
under brittle conditions. The deformation has produced joints, frac-
tures and folds (Raiverman et al., 1993; Srivastava and John, 1999).
The Piedmont Zone, south of the HFT, is constituted by coalescing
alluvial fans deposited by the streams with low gradient emerg-
ing from the Siwalik mountain range. The region receives about
1600 mm/yr rainfall of which about 80–85% falls during the months
of July to September. These streams originate in the Outer Himalayan
Foot Hills and drain to the Ganga plain. The stream beds are com-
posed of rounded pebble and cobble gravels in the sand matrix and
locally, within the gravel bars beds of well sorted fine grained sand.
The streams have often cut two or three sets of terraces. The fan ter-
races consist of unconsolidated gravels, sand and mud through which
streams have often cut deep gorges. The fans slope 5°–7° to south and
also support local drainage. Due to rapid changes in the terrain and
the fast progressing reclamation of land for agricultural activity and
construction of numerous water storage structures along the moun-
tain front, the original depositional features are often severely modi-
fied. The terraces observed along some of the streams consist of
gravel and sandy layers with local clay horizons. South of the HFT, a
number of possible blind faults and lineaments have also caused
local uplift and subsidence of the ground (Yeats and Thakur, 2008).
The uplifted terrain has undergone rapid erosion due to intense
rainfall resulting into gullying and progressive development of
badlands.
2. Fluvial terraces
Singhauli Nala is one among the many seasonal streams emerges
from the Siwalik range and traverses through the study area, which
has high discharge during the monsoon in the HFT zone. Three main
165
terraces (T1, T2 and T3) observed along the Singhauli Nala, which are
situated at 2 m, 5 m and 20 m above the present day stream bed
(Fig. 4). The oldest strath terrace (T3) deposit, consists of 2–3 m thick
rounded pebble–cobble–gravel bed with overlying 1–2 m thick layer
of loamy sand (Fig. 5) resting over the Siwalik rocks (15 m). However,
the younger terraces (T2 and T1) do not have any capping layer of
loamy sand or if existed must have eroded away in due course of
time. Explicit expression of the Middle Siwalik sandstone riding over,
uplifting and deforming the terrace material is observed at the exit of
Singhauli Nala, near Singhauli village. Here the uplifted strath terrace,
resting over the Middle Siwaliks is abruptly truncated by the HFT
(Fig. 6a). This terrace (T3) surface, which is ~20 m above the present
day river bed forming the hanging wall of the HFT, has also been
upwarped (Fig. 6b). At the same elevation of ~20 m, uplifted strath
terrace (~T3) are also observed at Pamuwala (Fig. 3a), 2.5 km west of
the trench site. The terrace is comprised of 2–3 m thick rounded
pebble–cobble–gravels and overlying 1–2 m thick layer of loamy sand
(Fig. 5) as in the case of the T3 terraces at Singhauli Nala. Fluvial strath
terraces are widely observed along the northwestern Frontal Himalaya.
In the Mohand region (Fig. 2), which is ~70 km east of the present study
area, similar fluvial strath terrace deposits exposed 20 to 30 m above
the present day stream level are observed (Wesnousky et al., 1999).
Comparable fluvial strath terraces of the Markanda River, which is
~8 km west of the present study area (Fig. 3c), placed 23 m above the
active stream are also observed (Kumar et al., 2001). These fluvial strath
terraces are interpreted to have been uplifted by the activity of the HFT.
3. Trench excavation and paleoseismic investigation
A trench excavation survey for paleoseismic investigation was car-
ried out ~ 11 km east of Kala Amb township, along the left bank of
Singhauli Nala, across the faulted and displaced fluvial terrace in the
topographic front of the HFT at Singhauli village (Fig. 7). The exposure
has been further excavated (32 m long and 7 m deep) for detailed
paleoseismic investigation. The hand-log prepared of the trench
wall (Fig. 8) helped reconstruction of the litho-stratigraphic succes-
sion of the deposits, which consist mainly of clastic deposits of fluvial
origin (Late Pleistocene and Holocene sediments) overlying the Mid-
dle Siwaliks (the Tertiary rocks). Although the lithologic units in the
trench wall show stratification, they are further deformed and the
sedimentary units at places have distinct erosional contacts with the
overlying units.
3.1. Litho-units in the trench
Based on variations of color, matrix, size and the distribution of
the clasts, it was possible to distinguish individual sedimentary
units within the excavated section. Besides the oldest Middle Siwalik
sandstone (Unit A), the Quaternary units are divided into eight sepa-
rate principal sedimentary sub lithological units, from Unit B (oldest)
to Unit I (youngest) (Fig. 8). They are comprised of sub angular to an-
gular, pebble to boulder clasts and are matrix or clasts-supported. The
boulders vary in size from few centimeters to at places > 1.5 m.
Compositionally the clasts consist of gray and pink sandstone indicat-
ing their derivation from the Tertiary mountain ranges. The thickness
of the individual sedimentary sub litho-units ranges between 20 cm
to more than 2 m.
Unit B, the oldest unit in the trench is composed silty sand as injec-
tion feature (Fig. 8). Unit C, consisting of silty clay with infrequent oc-
currence of pebbles and gravel, has a maximum thickness of 2 m. This
unit is dramatically deformed (folded and faulted) in a non-brittle man-
ner and at places its thickness also has changed severely. Unit D is com-
posed of broken sub angular boulders and pebbles with sandy clay
matrix besides occasional occurrence of few large sized boulders
whose ratio to matrix is less than 20%. This unit at some stage has also
undergone severe deformation and got dismembered, therefore its
166 G. Philip et al. / Tectonophysics 580 (2012) 162–177

Fig. 4. Fluvial terrace deposits: a. Panoramic view of the three levels of fluvial terraces in the area which are overlying the uplifted Middle Siwaliks. View is towards north. T3 is the
oldest and maximum uplifted terrace, whereas T2 and T1 are the younger terraces formed subsequently. b. Close up view of the area shown within the box in figure a. Left bank of
the Singhauli Nala shows the disposition of T3 and T2 terraces and the underlying Middle Siwalik sandstone.

thickness also varies. The trench wall shows that Units C and D are
emplaced and deformed by upward and lateral injection of fine-
grained materials (Unit B) during a tectonic strain-releasing event.
Unit E mainly consists of clayey sand (dry) with occasional presence
of pebbles within it. The thickness of this unit varies from few centime-
ters to about 2 m in the trench. This unit also shows deformation and its
subsequent erosion. Unit F consists of mainly gravels and pebbles and
appeared to have undergone deformation. Unit G is the most widely
distributed unit in the trench and composed mainly of large sized
rounded boulders, pebbles and gravels. Unit H, overlies Unit G, is mainly
composed of loamy sand. Its exposure is limited and gets mixed up with
the colluvium. Unit I is essentially a channel fill deposit, developed due
to partial incision of Units D, E, F and G, has also composed of large sized
unsorted boulders, pebbles and gravels. The clasts are predominant as
compared to the matrix along an erosional contact with underlying
Unit E.
3.2. Faults
and E. However, the tip of Fault I has been later on cut and covered
(Fig. 9a) by a channel fill deposit (Unit I).
Another prominent fault, Fault II, ~5 m north of Fault I, has been
identified in the trench (Figs. 7a, c, and 8). The fault plane shows a dip
of 30°N where the Middle Siwalik sandstones (Unit A) thrust over the
fluvial terrace deposits (Units G and H). Fault II has uplifted and
upwarped the fluvial terrace (T3). The base of T3 terrace is ~10 m
above the T2 terrace surface in the Singhauli Nala. Pebbles belonging
to the terrace deposits have also got entrained within the highly
sheared Middle Siwalik sandstone (Fig. 10a) in this fault zone. In
mudstone the closely spaced slaty cleavages developed only at the
base of the hanging wall of Fault II (Fig. 10b). Displacement of rocks
along the HFT has stretched the incompetent rocks due to
layer-parallel simple shearing. Extensional fractures (through which
seepage of water takes place) have also developed transverse to the
bedding-parallel cleavages. The exposed outcrops were left for further
degradation by ongoing surface fluvial processes. The external geome-
try of the lithounits, the Siwaliks and the fluvial terrace (T3) in the
hanging wall show upwarping.

From south to north, the trench wall shows three thrust faults: Fault
0, Fault I and Fault II, which are parallel to sub parallel to the thrust
plane of the HFT (Figs. 6–8). Towards southernmost part of the trench
within the Late Pleistocene fluvial sediments, Fault 0 displaced and de-
formed Units B, C, D, E and F. Towards ~9 m north of Fault 0 another
fault, Fault I is observed (Fig. 8). This fault (Figs. 7b and 8) has a dip
23°N where the Middle Siwalik sandstone has thrust over Units B, C, D
3.3. Induced deformation features — paleoliquefaction
Well defined diagnostic soft-sediment deformation and
paleoliquefaction features are recognized in the trench wall. These fea-
tures are irregular in their size and shape and are basically consist of
silty sand as injection features (Fig. 11). The large-sized sand injection
has not only vertically uplifted and upwarped the overlying lithounits
 G. Philip et al. / Tectonophysics 580 (2012) 162–177 167

Fig. 5. Fluvial strath terraces: a. Uplifted (20 m from the present day river bed) Quaternary fluvial terrace (T3) in the hanging wall at the HFT along Singhauli Nala. The terrace de-
posit has 1–2 m thick loamy sand at the top and 2–3 m thick gravel–pebble–boulder pebble–cobble–gravel below. The above assemblage overlies the Middle Siwalik sandstone at
Singhauli Nala. b. Another terrace with 3–5 m thick deposits observed at Pamuwala, 2.5 km west of Singhauli showing base of the terrace at 15 m above from the present day
Pamuwala Nala.

(Fig. 8) but has also dismembered some of the units (Units C and D). The
bowl shaped liquefied sand in the trench shows a vertical height of
~5 m and has a lateral spread up to ~9 m (Figs. 8 and 11a). Part of the
dismembered units is also found trapped within the large injection fea-
ture (Fig. 12). The sand injection feature is truncated by Fault I and has
neither dismembered nor deformed the overlying lithounits. Apart
from this, other irregular features due to the liquefaction and vibrations
originating from earthquakes or the water-escape from sediments are
also observed in the trench. Another relatively small and parallel exca-
vation was carried out at a distance of ~10 m close to the right bank of
Singhauli Nala away from the present liquefaction feature in the trench,
has also verified the lateral extension of the large dimensional induced
liquefaction feature (Fig. 11b).
the fluvial and strath terraces (Fig. 13). The Leica Total Station offers the
accuracy of angular measurement 1σ = 0.15 mgon (0.5″) and is equiv-
alent to a transverse deviation of 12 μm at a range of 5 m. Although the
fault scarp has been degraded to a greater extent today, the topograph-
ical cross sections (Fig. 13c–f) show discreet surface elevation variations
in the fault zone thereby substantiating manifestation of the surface de-
formation due to repeated faulting in the area. The topographic profiles
have also defined the elevations of important geomorphic marker sur-
faces with reference to the present day river bed elevation (~345 m)
of the Singhauli Nala. The top surface of the T3 terrace has been marked
at 365 m and the T2 terrace surface at 350 m elevation although the
thicknesses of the terraces are marginally varying.

4. Topographic profiling
In the mountain front region of the HFT at Kala Amb, the trench
shows multiple faulting events. To comprehend the surface manifesta-
tion and related ground surface elevation changes in the close vicinity
of the faults, few topographical cross profiles were prepared, using
Total Station (Leica TPS 5000) along and across the fault zone including
5. Luminescence chronology
Selected samples, representing most of the lithologic units in the
hanging wall and foot wall of the faults, were dated using the OSL
technique to constrain the chronology of events (Table 1). The quartz
grains from the samples were extracted after removing carbonate and
organic matter by chemical treatment (10% HCl and 30% H2O2), siev-
ing, density separation and HF etching (80 min) by following the
168 G. Philip et al. / Tectonophysics 580 (2012) 162–177

Fig. 6. Himalayan Frontal Thrust: a. Panoramic view of the mountain front and Himalayan Frontal Thrust (HFT) near Kala Amb where the Middle Siwalik rocks (MSw) are thrusting
over the Quaternary alluvium. View is towards the east. The trench location is shown within the box (outlined in red). b. The strath terrace (T3), 3 to 5 m thick Holocene sediments
resting on the exposed 15 m thick Middle Siwaliks, rides over Late Pleistocene sediments (LPs) along HFT. The T3 terrace is abruptly truncated by the HFT at the trench site. The
arrows (in green) show reactivation of the HFT along two discrete faults.

standard methods (Aitken, 1985). The purity of the etched quartz
(i.e. for any feldspar contaminations) is checked by infra-red stimulated
luminescence technique. The etched quartz grains were later fixed into
stainless steel disks (small aliquots of 3 mm size) using silicon oil for
determining the radiation energy received by the sample after its burial
(i.e. paleodose or equivalent dose). The OSL measurements were carried
out on an automated Riso TL-DA 20 reader equipped with blue light
emitting diodes following Single Aliquot Regeneration (SAR) protocol
(Murray and Wintle, 2000, 2003). The equivalent dose values were
calculated using the initial integral of the OSL (0.8 s) using Duller's
Annalyse software. For the annual dose rate estimation, concentrations
of uranium, thorium and potassium in the sediments were measured by
XRF. The present water content was determined for all the samples by
heating them at 100 °C.
Initially, eleven selected samples from the trench as well as from the
adjoining fluvial terraces were dated by the OSL technique (Table 1).
The two samples (SNT-8 and SNT-12) dated from Unit B, the in-
duced sand injection structure (Fig. 8) show ages of 79.25± 9.61 and
63.35 ±7.46 ka, respectively. The bottom most exposed portion of
Unit C (Fig. 8) in the trench (SNT-3) gives an age of 34.12± 4.49 ka.
Similarly the OSL samples from Unit C, in the hanging wall (SNT-17)
and foot wall (SNT-18) of Fault 0 also show comparable ages of
34.51 ±3.63 ka and 36.37± 2.42 ka, respectively. Unit E in the footwall
(SNT-13) of Fault I shows an age of 29.26± 1.92 ka. The sample
(SNT-24) from the top portion of the Late Pleistocene sediments (Unit
H) in the footwall of Fault II (Fig. 8) shows an age of 17.02±1.96 ka.
The top most strath terrace i.e. T3 terrace (Fig. 13b) gives an age
(SNQ-41) of 5.78± 0.33 ka. The channel fill deposit (SNQ-25) observed
 G. Philip et al. / Tectonophysics 580 (2012) 162–177 169

Fig. 7. Earthquake faults: a. Panoramic view of the trench wall (with 1 m grid) showing two distinct faults (F–F shown in arrows), where the Middle Siwalik sandstone (MSw)
thrust over the Late Pleistocene sediments (LPs) at the HFT, Kala Amb. b. and c. Close up views of part of the two paleoearthquake faults (Fault I, the older and Fault II, the younger)
observed in the trench. Depositional contact of the Quaternary alluvium with the Middle Siwaliks is shown by shaded lines.

in the trench (Fig. 8) has an age of 4.27 ±0.61 ka. The T2 terrace to-
wards south (Fig. 13b), adjoining the trench, shows a younger age of
2.06± 0.18 ka (SNQ-3) to the topmost litho-unit and the T1 terrace
(SNQ-4) shows the youngest age as 1.2± 0.02 ka (Fig. 13b).
6. Discussion
The HFT is manifested in the form of discontinuous range front scarps
that truncate the Quaternary fluvial terraces and alluvial fans (Nakata,
1972, 1989; Yeats et al., 1992). The major historical earthquakes in the
Himalaya have originated underneath the Higher Himalaya, north of
the surface trace of the MCT, and have ruptured the complete basal
decollement southward to the HFT (Seeber and Armbuster, 1984). The
two faults (Fault I and Fault II) observed in the trench wall at Singhauli
(Figs. 7 and 8), where the Middle Siwaliks have thrust over the Quaterna-
ry alluvium, are very significant as far as the repeated reactivations of the
HFT in the Frontal Himalaya are concerned. These two faults have not
only uplifted the younger fluvial terraces but also generated induced sec-
ondary features and soft-sediment deformation structures, mainly the
sand injection and water escape structures.
6.1. Paleoliquefaction and generation of Fault 0 and Fault I
Paleoliquefaction studies are very significant while investigating the
areas that were liquefied during large-magnitude paleoearthquakes
(Sims and Garvin, 1995; Tuttle and Seeber, 1991) and are more relevant
to the assessment of seismic hazards. After examining the worldwide
data on historical earthquakes it has now been established that
liquefaction effects become relatively common during an earthquake
of magnitude >5.5 (Ambraseys, 1988). Depths of liquefied source
beds most commonly reported are less than a few tens of meters
(Seed, 1979). The minimum value of peak accelerations required for
formation of liquefaction features decreases with increasing magnitude,
with reported values as low as 0.025 g for M 8.25 and 0.12 g for M 5.5
(Carter and Seed, 1988). Clastic intrusions (dykes and sills) in sedi-
ments are the major effects of seismic liquefaction, which are generally
observed at shallow depth, up to about 10 m deep, and typically involve
very significantly elevated pore water pressures. The paleoliquefaction
feature (Figs. 8 and 11) in the trench suggests the secondary effects of
a large magnitude paleoearthquake that must have occurred in the
area. The induced structures are formed where strong hydraulic forces
of short duration are suddenly applied such as sudden release of tecton-
ic strain during an earthquake resulting to liquefaction (Obermeier,
1996). During strong seismic shaking, water and sediment (generally
sand) flow from depth into fissures along the base of an overlying
fine-grained cap. Associated with liquefaction features, water escape
structures are also contemporaneously developed during the seismic
event. Since they are very commonly related to seismic activity, there-
fore are classified as secondary/induced features of large earthquakes
(McCalpin, 1996; Yeats et al., 1997). The mechanism of folding,
deformation and development of liquefaction and water escape struc-
tures hence have definite genetic link with the occurrence of a large-
magnitude earthquake.
The OSL ages of the samples (Table 1: SNT-8 and SNT-12) collected
from two different spots within the induced sand injection structure
(Unit B) show ages of 79.3 and 63.4 ka. The sand injection structure
 170
G. Philip et al. / Tectonophysics 580 (2012) 162–177
Fig. 8. Trench log: a. Field photograph showing the trench excavation site at Singhauli. The rectangle shows the area of trench log prepared. Fault 0, Fault I and Fault II are also marked in the photo. b. Close up photograph of Fault 0 in the
trench wall (marked in red box in the trench log). c. The trench log of the left bank of Singhauli Nala near Kala Amb showing repeated reactivation of the HFT in the mountain front. The different litho-units (A–I) are individually identified
based on the clast and matrix type and distribution. Two distinct faults recognizable in the trench log indicate their activities in the Late Pleistocene and Holocene times. The large-sized induced liquefaction feature (Unit B) and the deformed
and dismembered units are also observable in the trench log. The OSL sample locations (except for two, which are outside the trench) are shown with their ages.
 G. Philip et al. / Tectonophysics 580 (2012) 162–177 171

Fig. 9. Abandoned channel: a. Part of the trench wall showing an abandoned channel at the scarp front of the HFT. This channel has eroded Siwaliks, Fault I and underlying Late
Pleistocene deposits. b. Close up view of the abandoned channel showing the flow direction (shown in arrow), which has eventually eroded away the southern extension of
Fault 0 is also visible in the photograph.

has cut across Unit C suggesting the injection phenomenon occurred
after the deposition of Unit C (SNT-3, 34.1 ka). The sand injection has
also uplifted and pierced into the younger litho-units (D and E) in the
trench. Unit E (SNT-13) shows a younger age of 29.3 ka which implies
that the actual sand injection has happened only after the deposition
of Unit E. The dismembered units suggest that during heavy shaking,
the water escape structures have also altered the thickness of the sedi-
mentary units by upward and lateral injection of fine grained materials
(Fig. 8). The size of the injection feature and the caught up chunks (Unit
D) of the Late Pleistocene gravels (Fig. 12) corroborate its co-seismic oc-
currence with a large magnitude earthquake.
Fault I demonstrates the activity of the HFT by thrusting of Siwalik
rocks over the Late Pleistocene alluvium. After the rupture, the Middle
Siwaliks in the hanging wall remained under sub-aerial weathering and
erosion. The scarp associated with Fault I must have reached the surface,
however, was later on eroded away by fluvial activity. Subsequently Units
G and H were deposited by the then active Singhauli Nala. The tip of this
fault was cut and covered by Unit I, a channel fill deposit (Figs. 8 and 9a).
A parallel fault, Fault 0, is observed in the footwall of Fault I which
has affected Units C, D, E and F (Fig. 8). The samples (SNT-3 and
SNT-17) from the hanging wall of Fault 0 show ages of 34.1 and
34.5 ka, respectively and its counterpart in the foot wall (SNT-18)
shows a comparative age of 36.4 ka. This confirms the faulting event
and indicates that Fault 0 was generated after the deposition of Unit C.
Unit E (SNT-13) on the other hand shows a younger age of 29.3 ka.
Based on the OSL ages and disposition of Units C, D and E it is suggested
that Fault 0 must have generated after 29.3 ka i.e. after the deposition of
Unit E. With the help of the deformation pattern of the boundaries
between C, D, E, F, we estimate 2.5–3 m displacement along Fault 0.
Injection of the liquefied sand (Unit B) has also accommodated by the
folded and displaced Late Pleistocene units along Fault 0. The folding
pattern and disposition of the units show that this fault cuts (at least
strongly deforms) Units C, D, E and F but has unaffected Unit I. Hence
we believe that the timing of generation of Fault 0 is also time-
equivalent as that of Fault I.
It is observed that the units in the foot wall of Fault I do not have
their counterpart units in the hanging wall (Fig. 8). This suggests that
Units C, D, E and F had been at a higher level and were eroded away
before the deposition of Unit G. Hence there must have been at
least two faulting events along Fault I: one was after the deposition
172 G. Philip et al. / Tectonophysics 580 (2012) 162–177

deposited later on. Since no counterparts of the respective Quaternary

units of the footwall are observed on the hanging wall of Fault I, the
cumulative displacement along the fault plane is estimated to be
12 m or more.
It is also noticed that the paleoliquefaction features neither have
cut across Fault I nor dismembered the younger Units G, H and I.
The base of Unit G on the up thrown side, gently goes down to the
south, without upward deformation. If Fault I was activated after
the deposition of Unit G, the unit must have also been suffered defor-
mation. Therefore, we believe that Fault I activity and the associated
liquefaction must have occurred before the deposition of Unit G. An
OSL sample (SNT-24) from the upper part of Unit H, in the hanging
wall of Fault I shows an age of 17 ka. Hence by considering these
two ages of Units E (29.3 ka) and H (17 ka), we believe that the
reactivation of the HFT has generated a large magnitude earthquake
resulting to the development of Fault I, which occurred after 29.3 ka
but before 17 ka.

6.2. Generation of Fault II

The hanging wall of Fault I, comprising Middle Siwaliks and Units

Fig. 10. a. Pebbles (P) derived from Unit G entrained within the Middle Siwalik sand-
stone observed at the fault (for location refer Fig. 8c) zone b. The highly sheared Mid-
dle Siwalik sandstone in the HFT zone. At the base of the hanging wall of Fault II, the
Siwaliks show bedding parallel closely spaced slaty cleavages, which are transversed
by extensional fractures (EF). For location refer to Fig. 8a and c.
of Units C–F and another faulting event must have occurred at a
later stage with the time equivalent of Fault 0. Subsequently, the
units (C–F) in the hanging wall must have eroded away and Unit G
G and H, is again thrust over by the Middle Siwaliks (Unit A) along an-
other parallel fault, Fault II. The fault plane dips 30°N, which is also
parallel to average dip of the HFT in this region, suggesting it to be
yet another reactivation of the HFT (Figs. 7 and 8). Fault II has uplifted
the Holocene terrace (T3) in the hanging wall and truncated across
the HFT (Figs. 5 and 7). The closely spaced shear zones (Fig. 10b) sug-
gests intense deformation in the Siwalik rocks during the generation
of Fault II by an earthquake. The scarp face of Fault II is presently
being degraded and covered by the colluvium derived from Siwaliks
and the upthrown T3 terrace. Owing to the precariously placed and
highly fragile litho-units, besides the site falling in the ‘Protected
Forest’ area, we could not carry out any kind of further cleaning or
sampling of the fault scarp to clearly identify the counterpart of the
T3 terrace in the footwall of Fault II. In this situation a convincing pos-
sibility is that T3 terrace might be distributed just on Unit H being
covered by reworked materials. Subsequently this must have incised

Fig. 11. Liquefaction features: a. Part of the trench wall showing sand injection (Si) features. The arrows show the direction of the movement of sand from subsurface that has
spread laterally also. The dashed white line shows some of the curved stratification preserved within the injection feature. Caught up chunks of Unit D along with the injection
features are shown in thin dashed yellow lines. The sand injection has not only cut across the Late Pleistocene sediments (LPs) but also uplifted and dismembered other top
lying lithounits as well. The dashed line (in dark blue) shows thrusting of Middle Siwalik sandstone Unit A over the LPs. b. Sand injection feature observed in the parallel excavation
carried out at a distance of 10 m away from the trench showing lateral extend of the feature (for location refer Fig. 13b).
 G. Philip et al. / Tectonophysics 580 (2012) 162–177
Fig. 12. a. Close up view of part of the trench wall showing dismembered and caught-up chunk of Unit D within the sand injection Unit B. b. Close up showing a part of the sand
injection (Unit B) into Unit C and few caught up boulders (Bd) within the injection feature observed in the trench. The trench grid is also visible in the photograph. Outline of the
ball point pen is for scale.
by the channel (Unit I) and covered by the T2 terrace. Also, the youn-
ger T2 terrace in this region does not show any offsetting. Therefore,
the upper half of the Older terrace shown in profile D–D′ (Fig. 13f) is a
possible horizon or niche of the counterpart T3. The OSL sample
(SNQ-41) collected 1 m below the T3 terrace surface (Fig. 13b) in
the hanging wall of Fault II shows an age of 5.8 ka. The above age is
broadly comparable to the radiometric dates (4.9 and 5.2 ka) on
two fragments of charcoal collected from fluvial strath terrace sur-
faces, at 23 m elevation, of the Markanda River (Kumar et al., 2001)
and the age of strath terrace (3.8 ka) at 30 m elevation in the Mohand
region (Wesnousky et al., 1999).
Unit H in the footwall of Fault II shows an age of 17 ka. It is observed
that Unit H (=Older terrace) on the footwall of Fault II does not have its
counterpart on the hanging wall in the trench. This suggests that the
Older terrace had been at the same level as or higher than T3 terrace
and was eroded away before the deposition of T3 terrace materials.
Consequently, the vertical offset of the Older terrace is larger than that
of T3. Hence there must have been at least two faulting events on
Fault II: one is after the Older terrace deposition/before T3 terrace depo-
sition, and another faulting event occurred after the deposition T3 ter-
race. Thus the total displacement along Fault II is cumulative. In this
case we propose two possibilities/options to calculate the amount of
vertical displacement for Fault II.
The altitude of the base of T3 terrace deposit on the hanging wall is
about 360 m, while the altitude of Unit H in the trench (=Older terrace
in Fig. 13f) on the footwall is ~350 m. On the footwall, T3 terrace de-
posits must have been distributed above the Older terrace, therefore
the possible lowest altitude of the base of T3 terrace deposits is inferred
to be ~350 m. Consequently the vertical displacement of T3 terrace
should be about 10 m and the displacement along the fault will be
~20 m. Considering the two successive faulting events along Fault II, it
is possible to consider 10 m as the minimum cumulative vertical offset
for all the estimation towards slip rate and magnitude.
On the other hand, Unit G and the lower portion of T3 terrace are
compositionally comparable since both the units mainly consist of
large sized rounded boulders, pebbles and gravels. Similarly Unit H
is also comparable with the upper portion of the T3 terrace because
both the units are composed of loamy sand. However, the OSL age of
Unit H (17 ka) is older than the upper portion of the present T3 terrace
(5.8 ka). On the footwall, the possible lowest altitude of the base of
comparable Unit H deposits is at ~ 349 m. Therefore occurrence of at
least two faulting events along Fault II must have resulted to the pres-
ent day elevations of Unit H and the T3 terrace. In this case if we ignore
173
the OSL age of Unit H and consider units G and H (= the present Older
terrace) as T3 terrace or its equivalent, in such supposition the vertical
offset of T3 terrace becomes about 11 m (360 m − 349 m = 11 m).
Hence the cumulative maximum vertical offset along Fault II due to
multiple events should be about ~ 11 m and the displacement will be
~ 22 m.
We propose two probable constraints on the timing of the last
movement along Fault II. A tighter one is before the deposition Unit I
and a looser one is before the formation of T2 terrace. The first possibil-
ity is based on the closeness of the OSL ages of T3 (SNQ-41) 5.8 ka and
Unit I (SNQ-25) 4.3 ka. Since the age of deposition is almost the same
for the above two units (T3 and Unit I), we reason here that the faulting
event must have occurred between 5.8 ka and 4.3 ka. On the other hand
the T2 terrace, both in the hanging wall and foot walls observed to be
unaffected by the faulting. Another likely possibility suggests that all
the surface deformation of older terraces including the last event on
Fault II predates the deposition of T2 terrace. Hence in general it may
be stated that last phase of faulting along Fault II must have occurred
in this area between 5.8 ka and 2 ka.
Based on the above observation, we report here for the first time
explicit evidences for multiple paleoearthquakes during Late Pleisto-
cene and Holocene in the Kala Amb area of the northwestern part of
the Frontal Himalaya. The topographical profiles in the present site
suggest that the distribution of T2 terrace (2.1 ka) along the river is
not affected by the movement or offset by the HFT. This suggests
that after 2 ka no tectonic activity is observed or identified in this
locality.
6.3. Uplift, slip and convergence rate along HFT at Kala Amb
The recent geodetic, GPS and seismic studies in the Himalaya have
provided significant data on the ongoing crustal deformation in the
Holocene time on a regional scale (Banerjee and Bürgmann, 2002;
Chander, 1988). The GPS based measurements show that the India–
Eurasia ongoing convergence rate is ~ 37–50 mm/yr (Banerjee and
Bürgmann, 2002; Bilham, 2004; Bilham and Wallace, 2005; De Mets
et al., 1994; Paul et al., 2001) and only ~ 10–20 mm/yr is accommo-
dated along the HFT (Bilham et al., 1997; Lave and Avouac, 2000;
Wesnousky et al., 1999). Although the MCT and MBT are considered
to be the major tectonic features of the Cenozoic shortening along
the entire extent of the Himalaya, the HFT, bordering the Himalayan
mountain front as well as delineating the northern limit of the ex-
posed Indian plate, is believed to be the most active thrust during
174 G. Philip et al. / Tectonophysics 580 (2012) 162–177
the Quaternary period (Kumar et al., 2006; Philip and Virdi, 2006;
Yeats and Lillie, 1991; Yeats et al., 1992).
Based on paleoseismological studies, the long term uplift and slip
rates have also been reported from the northwestern part of the
Himalaya (Kumar et al., 2006; Malik et al., 2010; Wesnousky et al.,
1999). Longer term uplift along the HFT is reflected in fluvial strath
terraces in the Frontal Himalaya. The vertical uplift rate, calculated
using one radio carbon age (3.8 ka) from a 20–30 m thick fluvial strath
terrace in the HFT region in the Mohand area was estimated as >6.9±
1.8 mm/yr besides the long term slip rate of the HFT to be ≥13.8 ±
3.6 mm/yr (Wesnousky et al., 1999). The uplifted Markanda River
terrace, which is about 23 m height, was dated using two charcoal sam-
ples. The radiometric dates on two fragments of charcoal provide a
maximum age of the terrace surface: 4.9 ka and 5.2 ka, respectively.
Based on this, the uplift due to slip on the HFT gives 4.8 ± 0.9 mm/yr
(Kumar et al., 2001). Mugnier et al. (2004) estimated kinematics and
rates of displacement along single fault of the Himalayan Frontal Belt
in western Nepal on the basis of balanced cross sections, fluviatile
terrace-deposit, and geodesy. The convergence rate between the Indian
and Tibetan plates has been estimated to be consistent (19–20 mm/yr)
for the past 10 million years.
In the present study, the uplifted T3 strath terrace (OSL age 5.8 ka) is
placed almost at a height of 15 m from the present day river bed. The
distribution of the fluvial terraces in these areas and their near similar
age of deposition suggest the longer term uplift along the HFT. The ele-
vation of the fluvial strath terraces is used to calculate the slip and the
uplift rate of the HFT at Singhauli. The T3 Terrace, 3–5 m thick and rest-
ing over the 15 m high Siwalik rocks, shows a long term uplift to be
3.4 mm/yr and has generated a slip of 6.8 mm/yr along the 30° dipping
Fault II. The large slip of 20–22 m along Fault II might also due to mul-
tiple reactivations along the fault.
6.4. Magnitude
The earlier paleoseismological studies in the Frontal Himalaya
(Kumar et al., 2001, 2006; Malik et al., 2003, 2010; Philip et al., 2011;
Wesnousky et al., 1999) suggest that there have been 8.2 b Mb 8.7
earthquakes, larger than either the 1905 Kangra or the Bihar–Nepal
1934 ruptures (Bilham and Wallace, 2005). The paleoseismological
studies along the HFT in the southeastern Nepal where 4 m displace-
ment observed in the trench is estimated to have occurred ~1.2 ka
(Upreiti et al., 2000). Meter-scale episodic displacements along the
HFT in western Nepal has also been studied (Mugnier et al., 2005)
based on structural and morphological evidences. In the hanging wall
of the HFT, encased fluvial strath terraces observed are interpreted to
be due to sudden uplift events. The abandonment of these terraces are
believed to be linked to 2 m slip events along the HFT, a value close to
the lower bound inferred for co-seismic slip along the basal detachment
during Mw 8 earthquakes. From the offset of fluvial terrace level, a dis-
placement of ~8 m is inferred between 1.2–1.3 ka and 1.8–1.9 ka
(Mugnier et al., 2005).
In the northeastern Himalaya, along the HFT within the Indian
states of West Bengal and Assam, Kumar et al. (2010) have
conducted paleoseismic studies between the meizoseismal areas
of the 1934 Bihar–Nepal and 1950 Assam earthquakes. Of these,
two sites show that offsets during the last surface rupture event
were at minimum of about 14 m and 12 m. However, the scatter
in radiocarbon ages on detrital charcoal has not resulted in a firm
bracket on the timing of events observed in the trenches. On the
other hand the offset is relatively much smaller (b2.5 m) in third
Fig. 13. Topographical profiles: a. Topographic map of the area around Singhauli near Kala Amb. b. Close up of the box in Fig. 13a showing the landform map of the area around the
trench site at Singhauli. Different levels of terraces (T1–T3), abandoned course of Singhauli Nala (shown in dotted arrows), Active landslide (As), Denudo-structural hills (DsH) of
the Middle Siwaliks and the Piedmont alluvium (PdA). c–f. Topographic cross profiles showing variation in ground surface due to faulting disposition of the fluvial terraces and
degraded fault scarp (Fig. 13f) in the trench location. The sections except the profiles are approximate and not to the scale.
site and radiocarbon dating shows it to have occurred after A.D.
1100 (after about A.D. 1270).
The faults observed at Singhauli suggest that there must have
been large magnitude multiple paleoearthquakes along the HFT
during Late Pleistocene and Holocene. Empirical formulae are
being employed to calculate the parameters of paleoearthquakes.
Comparing source parameters of the historical earthquakes occurred
worldwide, Wells and Coppersmith (1994) suggested an empirical
relationship, related to earthquake magnitude (M), maximum
displacement (MD) and surface rupture length (SRL) as M =
6.69 + 0.74 ∗ log(MD) and M = 5.08 + 1.16 ∗ log(SRL). In the present
study for Fault I, no counterparts of the respective litho-units on
the downthrown side are observed on the upthrown side. However,
the measured slip of Siwalik rocks along the fault plane in the trench
is ~ 12 m. This fault must have caused a minimum vertical displace-
ment of 4.7 m along the 23° dipping fault plane. On the other hand,
based on boundaries of the corresponding litho-units, we measured
~ 4 m slip along the 23° dipping Fault 0 resulting to 1.6 m vertical
displacement. We therefore believe that a large earthquake with
~ 12 to 16 m of slip with a magnitude 7.5 or greater hit this region
in the period between 29 ka and 17 ka.
With regard to Fault II, considering the OSL ages of the T3 terrace
and Unit H besides comparative elevation of the base of Unit H and
upper part of the T3 terrace, two possibilities of vertical displace-
ment have been taken into account for estimating the magnitude.
Hence for the minimum vertical offset of 10 m, the magnitude will
be 7.6 whereas for 11 m off set the magnitude estimated to be
~ 7.7. We therefore believe that a large earthquake of magnitude
~ 7.7 or greater must have occurred in this region with a maximum
SRL of about 150 km. However, the observable SRL might vary
depending on lithology and rates of erosion. In the present study
the observed length of the subdued fault scarp is limited owing to
fast modification of the landform for agricultural and industrial
development.
On the other hand for Japanese inland earthquakes the magnitude
(M) is calculated using the formula logD = 0.6 M − 4.0, where D is the
displacement during the earthquake (Matsuda, 1975). Based on the
above relationship, the fault displacement observed at the Singhauli
trench indicates that an earthquake of magnitude of ~ 7.8 must have
occurred in the region during the generation of Fault I where as an
earthquake of magnitude 8.3 or 8.4 must have occurred during the
generation of Fault II. Considering the above two empirical equations
suggested, we are of the opinion that the reactivation along the HFT
has generated large paleoearthquakes of >7.5 M to 8.4 M during
Late Pleistocene and Late Holocene.
Few trenches excavated across the HFT for paleoseismological
studies (Kumar et al., 2001, 2006; Malik et al., 2003, 2008, 2010)
have also substantiated that the HFT along the mountain front be-
tween Chandigarh and Kala Amb, Adh Badri even at Hajipur ruptured
during the last 2000 years and generated two or three major earth-
quakes (M ≅ 7or 8). The 1.5 m slip observed in the reactivated
Nalagarh Thrust suggests major Late Pleistocene earthquakes in the
Outer Himalaya (Philip et al., 2011). The > 1.5 m slip along the PGF
(Malik et al., 2003) and the two earthquakes reported in the HFT re-
gion (Kumar et al., 2001) suggest that active faults in the Frontal
Himalaya are capable of generating large magnitude earthquakes.
However, the available historical data do not indicate occurrence
of any great earthquakes in the Garhwal–Kumaun Province, prior to
the 1803 earthquake, at least for a minimum period of 1000 years
(Rajendran and Rajendran, 2005). In this context given the
large uncertainties in the location and magnitude, few researchers
G. Philip et al. / Tectonophysics 580 (2012) 162–177 175
176 G. Philip et al. / Tectonophysics 580 (2012) 162–177

Table 1
Optically stimulated luminescence (OSL) ages of samples from the Trench and the Quaternary deposit at Singhauli. Elemental concentration of Uranium (U), Thorium (Th), and
Potassium (K), and moisture content used for dose rate calculation and equivalent dose are also given.

Lab no. Sample No. Unit in the Trench Depth from top (m) U (ppm) Th (ppm) Potassium K (%) Moisture Equivalent Dose rate Age (ka)
content (%) Dose (De) Gy (Gy/ka)

LD1152 SNQ-4 T1 Terrace 1 1.42 ± 0.01 5.4 ± 0.05 0.94 ± .01 1.00 2.20 ± 0.35 1.8 ± 0.02 1.2 ± 0.02
LD0753 SNQ-3 T2 Terrace 1 3.54 ± 0.04 13.5 ± 0.14 0.94 ± 0.01 0.63 5.79 ± 0.50 2.81 ± 0.04 2.06 ± 0.18
LD0528 SNQ-41 T3 Terrace 1 4.1 ± 0.04 10.4 ± 0.10 1.54 ± 0.02 4.46 18.39 ± 1.01 3.17 ± 0.04 5.78 ± 0.33
LD0754 SNQ-25 I 1.5 3.07 ± 0.03 10.2 ± 0.10 1.08 ± 0.01 0.51 11.12 ± 1.58 2.61 ± 0.03 4.27 ± 0.61
LD0752 SNT-24 H 3 3.28 ± 0.03 15 ± 0.15 1.47 ± 0.01 0.80 56.37 ± 6.46 3.31 ± 0.04 17.02 ± 1.96
LD1085 SNT-13 E 2 4.15 ± 0.4 10.7 ± 0.11 1.67 ± 0.02 0.46 100.90 ± 6.50 3.45 ± 0.04 29.26 ± 1.92
LD1086 SNT-17 C 3 3.28 ± 0.03 14.3 ± 0.14 2.16 ± 0.02 5.68 128.39 ± 13.41 3.72 ± 0.05 34.51 ± 3.63
LD0749 SNT-3 C 1 2.8 ± 0.03 17 ± 0.17 1.46 ± 0.01 8.32 106.06 ± 13.88 3.11 ± 0.05 34.12 ± 4.49
LD1093 SNT-18 C 4 0.93 ± 0.01 16.2 ± 0.16 2.26 ± 0.02 4.74 125.15 ± 8.18 3.44 ± 0.04 36.37 ± 2.42
LD0751 SNT-12 B 3 1.47 ± 0.01 9 ± 0.09 1.00 ± 0.01 1.75 128.09 ± 15.02 2.02 ± 0.02 63.35 ± 7.46
LD0750 SNT-8 B 3 1.42 ± 0.01 4.6 ± 0.05 0.95 ± 0.01 1.09 133.38 ± 16.11 1.68 ± 0.02 79.25 ± 9.61

(Ambraseys and Bilham, 2000; Arora et al., 2012; Gahalaut, 2008;
Kayal, 2010; Rajendran and Rajendran, 2005) have reserva-
tions about some of the recent interpretations of shorter recur-
rence of large/great earthquakes along the HFT reported from
paleoseismological studies (Kumar et al., 2001, 2006). These
studies and many new exposures, observed along new water
storage structures constructed along the frontal part, clearly re-
vealed that the Siwaliks (Mio-Pleistocene) are riding over the
Quaternary fans and terraces (Fig. 6) indicating not only the ac-
tive nature of the HFT but also that rupture is propagating to the
surface.
7. Conclusions
The trench excavation survey conducted across the HFT in the
Kala Amb area exhibits three thrust faults, Fault 0, Fault I and Fault
II, of which Fault I and Fault 0 are time equivalent. The two faults
(Fault I and Fault II) observed in the trench wall at Singhauli where
the Middle Siwaliks have thrust over the Quaternary alluvium are
very significant as far as the repeated reactivation of the HFT is
concerned. We believe that a large earthquake of ~ 12 to 16 m slip
with a magnitude 7.5 or greater hit the Himalayan Front during the
period between 29 ka and 17 ka resulted development of Fault
0 and Fault I. The hanging wall of Fault I, comprising Middle Siwaliks
and Late Pleistocene–Holocene sediments, is thrust over by the Mid-
dle Siwaliks and Holocene (T3 terrace) along Fault II, which suggests
yet another reactivation of the HFT in the Frontal Himalaya between
5.8 ka and 2 ka. The large slip of 20–22 m along Fault II might also
due to multiple reactivations along the fault. After 2 ka no tectonic
activity is observed or identified in this locality. The faults observed
at Singhauli suggest that there must have been large magnitude
multiple paleoearthquakes along the HFT during Late Pleistocene–
Holocene. We therefore believe that a large earthquake of magnitude
~ 7.7 or greater must have occurred in this region with a maximum
SRL of about 150 km. The present study strongly suggests that the
Himalayan Front will be hit by large magnitude earthquakes greater
than those events took place in the historical past.
Acknowledgments
We are exceedingly grateful to Dr. Yuichi Sugiyama, Geological Survey
of Japan, for his critical and constructive suggestions which have im-
proved the manuscript to its present form. We are also grateful to the
two reviewers, Prof. Jesus Galindo-Zaldivar, and the anonymous reviewer,
whose valuable comments helped us in improving the quality of the man-
uscript. Thanks are also to the Conservator of Forest and the Divisional
Forest Officer, Nahan Circle, Himachal Pradesh Forest Department and
the Village Pradhan at Singhauli for their help and cooperation during
the trench excavation survey. We are grateful to the Director, Wadia
Institute of Himalayan Geology, Dehradun for providing the facilities
and permission to publish this research work.
References
Aitken, M.J., 1985. Thermo-luminescence Dating. Academic Press, London, p. 359.
Ambraseys, N.N., 1988. Engineering seismology: earthquake engineering and structural dy-
namics. Journal of the International Association for Earthquake Engineering 17, 1–105.
Ambraseys, N.N., Bilham, R., 2000. A note on the Kangra Ms = 7.8 earthquake of 4 April
1905. Current Science 79, 45–50.
Ambraseys, N.N., Douglas, J., 2004. Magnitude calibration of north Indian earthquakes.
J. Geophys. Int. 159, 165–206.
Ambraseys, N.N., Jackson, D., 2003. A note on early earthquakes in northern India and
southern Tibet. Current Science 84, 570–582.
Arora, B.R., Gahalaut, V.K., Kumar, N., 2012. Structural control on along-strike variation in
the seismicity of the northwest Himalaya. Journal of Asian Earth Sciences 57, 15–24.
Banerjee, P., Bürgmann, R., 2002. Convergence across the northwest Himalaya from
GPS measurements. Geophysical Research Letters 29 (13). http://dx.doi.org/
10.1029/2002GL015184.
Bilham, R., 2004. Earthquakes in India and the Himalaya: tectonics, geodesy and histo-
ry. Annals of Geophysics 47 (2), 839–858.
Bilham, R., Wallace, K., 2005. Future MW >8 earthquakes in the Himalaya: implications
from the 26 Dec 2004 MW = 9.0 earthquake on India's eastern plate margin. Geo-
logical Survey of India, Special Publication, 85, 1–14.
Bilham, R., Larson, K., Freymueller, J., Project Idylhim members, 1997. GPS measure-
ments of present-day convergence across the Nepal Himalaya. Nature (London)
386, 61–64.
Bilham, R., Gaur, V.K., Molnar, P., 2001. Himalayan seismic hazard. Science 293, 1442–1444.
Brown, L.D., Zhao, W., Nelson, K.D., Hauck, M., Alsdorf, D., Ross, A., Cogan, M., Clark, M.,
Liu, X., Che, J., 1996. Bright spots, structure, and magmatism in Southern Tibet from
INDEPTH seismic reflection profiling. Science 274 (5293), 1688–1690.
Carter, D.P., Seed, H.B., 1988. Liquefaction Potential of Sand Deposits Under Low
Levels of Excitation: Report No. UCB/EERC-81/11. College of Engineering, Univ. of
California, Berkeley. 119 pp.
Chander, R., 1988. Interpretation of observed ground level changes due to the Kangra
earthquake northwest Himalaya. Tectonophysics 149, 289–298.
De Mets, C., Gordon, R.C., Argus, D.F., Stein, S., 1994. Effect of recent revisions to the
geomagnetic reversal time scale on estimates of current plate motions. Geophysi-
cal Research Letters 21 (20), 2191–2194.
Gahalaut, V.K., 2008. Major and great earthquakes and seismic gaps in the Himalayan
arc. Golden Jubilee Memoir of the Geological Society of India 66, 373–393.
Gansser, A., 1964. Geology of the Himalaya. Interscience, New York, NY, p. 298.
Kayal, J.R., 2001. Microearthquake activity in some parts of the Himalaya and the tec-
tonic model. Tectonophysics 339, 331–351.
Kayal, J.R., 2010. Himalayan tectonic model and great earthquakes: an appraisal.
Geomatics, Natural Hazards and Risk 1, 51–67.
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 pattern in the Himalayan plate boundary and
identification of the areas of high seismic potential. Tectonophysics 96, 281–297.
Kumar, S., Wesnousky, S., Rockwell, G., Ragnon, D., T.K., 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., Rockwell, G., T. K., Briggs, R.W., Thakur, V.C., Jayangondaperumal,
R., 2006. Paleoseismic evidence of great surface rupture earthquakes along the Indian
Himalaya. Journal of Geophysical Research 111, B03304. http://dx.doi.org/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. Journal
of Geophysical Research 115, B12422. http://dx.doi.org/10.1029/2009JB006789.
Lave, J., Avouac, J.P., 2000. Active folding of fluvial terraces across the Siwalik Hills,
Himalayas of central Nepal. Journal of Geophysical Research 105, 5735–5770.
 G. Philip et al. / Tectonophysics 580 (2012) 162–177
Lyon-Caen, H., Molnar, P., 1983. Constraints on the structure of the Himalaya from the
analysis of gravity anomalies and a flexural model of the lithosphere. Journal of
Geophysical Research 88, 8171–8191.
Malik, J.N., Mathew, G., 2005. Evidence of paleoearthquakes from trench investigations
across Pinjaur Garden fault in Pinjaur Dun, NW Himalaya. Journal of Earth System
Science 114 (4), 387–400.
Malik, J.N., Nakata, T., 2003. Active faults and related Late Quaternary deformation
along the northwestern Himalayan Frontal Zone. Annals of Geophysics 46 (5),
917–936.
Malik, J.N., Nakata, T., Philip, G., Virdi, N.S., 2003. Preliminary observations from trench
near Chandigarh, NW Himalaya and their bearing on active faulting. Current Science
85 (12), 1793–1799.
Malik, J.N., Nakata, T., Philip, G., Suresh, N., Virdi, N.S., 2008. Evidence of a large magni-
tude earthquake along Chandigarh Fault in the frontal Himalayan zone, NW India.
Journal of Himalayan Geology 29 (2), 109–117.
Malik, J.N., Sahoo, A.K., Shah, A.A., Shinde, D.P., Juyal, N., Singhvi, A.K., 2010.
Paleoseismic evidence from trench investigation along Hajipur Fault, Himalayan
Frontal Thrust, NW Himalaya: implications of the faulting pattern on landscape
evolution and seismic hazard. Journal of Structural Geology 32, 350–361.
Matsuda, T., 1975. Magnitude and recurrence interval of earthquakes from a fault. Journal
of the Seismological Society of Japan 28, 269–283 (2nd Series).
McCalpin, J.P. (Ed.), 1996. Paleoseismology. Academic Press, New York. 588 pp.
Mugnier, J.-L., Huyghe, P., Leturmy, P., Jouanne, F., 2004. Episodicity and rates of thrust-
sheet motion in the Himalayas (western Nepal). In: McClay, K.R. (Ed.), Thrust
Tectonics and Hydrocarbon Systems AAPG Memoir 82, 91–114.
Mugnier, J.L., Huyghe, P., Gajurel, A.P., BecelMugnier, D., 2005. Frontal and piggy-back
seismic ruptures in the external thrust belt of Western Nepal. Journal of Asian
Earth Sciences 25, 707–717.
Murray, A.S., Wintle, A.G., 2000. Luminescence dating of quartz using an improved
single-aliquot regenerative-dose protocol. Radiation Measurements 32, 57–73.
Murray, A.S., Wintle, A.G., 2003. The single-aliquot regenerative-dose protocol: poten-
tial for improvements in reliability. Radiation Measurements 37, 377–381.
Nakata, T., 1972. Geomorphic history and crustal movements of foothills of the
Himalayas. Science Reports of the Tohuku, University, 7th Series: Geography,
22(1), p. 177.
Nakata, T., 1989. Active faults of the Himalaya of India and Nepal. Geological Society of
America Special Paper 232, 243–264.
Nelson, K.D., Project INDEPTH Team, 1998. The Himalaya and Tibetan Plateau: A
Perspective from Project INDEPTH. Paper Presented at 1998 Annual Meeting. Geo-
logical Society of America, Boulder, United States, p. 244.
Obermeier, S.F., 1996. Use of liquefaction-induced features for paleoseismic analysis —
an overview of how seismic liquefaction features can be distinguished from other
features and how their regional distribution and properties of source sediment can
be used to infer the location and strength of Holocene paleo-earthquakes.
Engineering Geology 44, 1–76.
Paul, J., Bürgmann, R., Gaur, V.K., Bilham, R., Larson, K.M., Ananda, M.B., Jade, S., Mukal,
M., Anupama, T.S., Satyal, G., Kumar, D., 2001. The motion and active deformation
of India. Geophysical Research Letters 28 (4), 647–650. http://dx.doi.org/10.1029/
2000GL011832.
Philip, G., Virdi, N.S., 2006. Co-existing compressional and extensional regimes along
the Himalayan Front vis-à-vis active faults near Singhauli, Haryana, India. Current
Science 90 (9), 1267–1271.
177
Philip, G., Suresh, N., Bhakuni, S.S., Gupta, V., 2011. Paleoseismic investigation along
Nalagarh Thrust: evidence of Late Pleistocene earthquake in Pinjaur Dun, north-
western Sub Himalaya, India. Journal of Asian Earth Sciences 40, 1056–1067.
Quittmeyer, R.C., Jacob, K.K., 1979. Historical and modern seismicity of Pakistan,
Afghanistan, Northwestern India, and southern Iran. Bulletin of the Seismological
Society of America 69 (3), 773–823.
Raiverman, V., Kunte, S.V., Mukherjee, A., 1983. Basin geometry, Cenozoic sedimenta-
tion and hydrocarbon prospects in north-western Himalaya and Indo–Ganga
plains. Petroleum Asia Journal 6, 67–92.
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.
Raiverman, V., Srivastava, A.K., Prasad, D.N., 1993. On the foothill thrust of northwest-
ern Himalaya. Journal of Himalayan Geology 4, 237–256.
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 Ser., vol. 4. AGU, Washington,
D.C., pp. 259–277.
Seeber, L., Armbuster, J., 1984. Some elements of continental subduction along the
Himalayan Front. Tectonophysics 105, 263–278.
Seed, H.B., 1979. Soil liquefaction and cyclic mobility for level ground during earth-
quakes. ASCE Journal of Geotechnical Engineering 105 (2), 201–255.
Sims, J.D., Garvin, C.D., 1995. Recurrent liquefaction induced by the 1989 Loma Prieta
earthquake and 1990 and 1991 aftershocks: implications for paleoseismicity studies.
Bulletin of the Seismological Society of America 85 (1), 51–65.
Srivastava, D.C., John, G., 1999. Deformation in the Himalayan Frontal Fault zone:
evidence from small scale structures in Mohand–Khara area, NW Himalaya. In:
Jain, A.K., Manickavasagam, R.M. (Eds.), Geodynamics of the NW Himalaya:
Gondwana Res. Group Memoir, 6, pp. 273–284.
Srivastava, J.P., Verma, S.N., Joshi, V.K., Verma, B.C., Arora, R.K., 1981. GSI Proceedings Neo-
gene/Quaternary Boundary Field Conference India. Geol. Surv. India, pp. 233–241.
Tuttle, M.P., Seeber, L., 1991. Historic and prehistoric earthquake-induced liquefaction
in Newbury, Massachusetts. Geology 19, 594–597.
Upreiti, B.N., Nakata, T., Kumhara, Y., Yagi, H., Okumura, K., Rockwell, T.K., Virdi, N.S.,
Maemoku, H., 2000. The latest active faulting in southeast Nepal. Proceedings of
the Hokudan International Symposium and School on Active Faulting, HOKUDAN-
2000, pp. 533–536.
Valdiya, K.S., 2003. Reactivation of Himalayan Frontal Fault: implications. Current
Science 85 (7), 1013–1040.
Wells, D.L., Coppersmith, K.J., 1994. New empirical relationships among magnitude,
rupture length, rupture width, rupture area, and surface displacement. Bulletin of
the Seismological Society of America 84, 974–1002.
Wesnousky, S.G., Kumar, S., Mohindra, R., Thakur, V.C., 1999. Uplift and convergence
along the Himalayan Frontal Thrust. Tectonics 18 (6), 967–976.
Yeats, R.S., Lillie, R.J., 1991. Contemporary tectonics of the Himalayan Frontal Fault
System: folds, blind thrusts and the 1905 Kangra earthquake. Journal of Structural
Geology 13, 215–225.
Yeats, R.S., Thakur, V.C., 2008. Active faulting south of the Himalayan Front: establishing a
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. The
Himalayan Frontal Fault system. Annales Tectonicae 6, 85–98 (Suppl.).
Yeats, R.S., Sieh, K., Allen, C.R., 1997. The Geology of Earthquakes. Oxford University Press.