Journal of Asian Earth Sciences 17 (1999) 703±712

Seismotectonics of the Nepal Himalaya from a local seismic

network
M.R. Pandey a, R.P. Tandukar a, J.P. Avouac b,*, J. Vergne b, Th. HeÂritier b
a
National Seismological Centre, Department of Mines and Geology, Kathmandu, Nepal
b
Laboratoire de GeÂophysique, CEA, BruyeÁres-le-ChaÃtel, France

Abstract

The National Seismological Network of Nepal consists of 17 short period seismic stations operated since 1994. It provides an
exceptional view of the microseismic activity over nearly one third of the Himalayan arc, including the only segment, between
longitudes 788E and 858E, that has not produced any M > 8 earthquakes over the last century. It shows a belt of seismicity that
follows approximately the front of the Higher Himalaya with most of the seismic moment being released at depths between 10
and 20 km. This belt of seismicity is interpreted to re¯ect interseismic stress accumulation in the upper crust associated with
creep in the lower crust beneath the Higher Himalaya. The seismic activity is more intense around 828E in Far-Western Nepal
and around 878E in Eastern Nepal. Western Nepal, between 82.5 and 858E, is characterized by a particularly low level of
seismic activity. We propose that these lateral variations are related to segmentation of the Main Himalayan Thrust Fault. The
major junctions between the di€erent segments would thus lie at about 878E and 828E with possibly an intermediate one at
about 858E. These junctions seem to coincide with some of the active normal faults in Southern Tibet. Lateral variation of
seismic activity is also found to correlate with lateral variations of geological structures suggesting that segmentation is a long-
lived feature. We infer four 250±400 km long segments that could produce earthquakes comparable to the M = 8.4 Bihar±Nepal
earthquake that struck eastern Nepal in 1934. Assuming the model of the characteristic earthquake, the recurrence interval
between two such earthquakes on a given segment is between 130 and 260 years. # 1999 Elsevier Science Ltd. All rights
reserved.

1. Introduction (Chander, 1988). The 1905 earthquake thus probably

ruptured a roughly 280 km long segment, from
The territory of Nepal spans about one third of the Kangra to Dehra Dun, that must have extended east-
length of the Himalaya (Fig. 1). Over the last century, ward to about 788E, near the border with Nepal
the Himalayan arc has experienced four earthquakes (Chander, 1988; Yeats and Lillie, 1991; Gahalaut and
with magnitude around 8.5 in 1897, 1905, 1934 and Chander, 1997) (Fig. 1). The 1934 Bihar±Nepal earth-
1950 (e.g., Seeber and Armbuster, 1981). The central quake probably ruptured a 200±300 km long segment
Himalaya of Nepal has not been active during this to the east of Kathmandu (Pandey and Molnar, 1988;
period (Fig. 1). The 1905 Kangra earthquake produced Molnar and Pandey, 1989) (Fig. 1).
severe damage in the Kangra area and, about 100 km Historical records indicate that the Kathmandu val-
to the east, in the Dehra Dun area (Middlemiss, 1910). ley has experienced recurring large earthquakes in the
Although site e€ects might in part account for the
past centuries. Major damage of probable seismic ori-
high intensities near Dehra Dun, comparison of level-
gin are reported to have occurred in 1255, 1408, 1681,
ing data shows coseismic elevation changes in this area
1803, 1810, 1833 and 1866 (Chitrakar and Pandey,
1986). Some of these events might be related to the
* Corresponding author. Tel.: +33-1-69-26-54-71; Fax: +33-1-69- repetition in the past of the 1934 Bihar±Nepal earth-
26-70-23. quake. That might be the case in particular for the

1367-9120/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S 1 3 6 7 - 9 1 2 0 ( 9 9 ) 0 0 0 3 4 - 6
704 M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712

Fig. 1. Distribution of probable rupture zones of the 1897, 1905, 1934 and 1950 earthquakes along the Himalayan arc. Modi®ed from Yeats and
Lillie (1991) and Yeats et al. (1992). Velocity of India relative to Eurasia computed from DeMets et al. (1990).

Fig. 2. Location of the short period telemetered stations of the National Seismological Network of Nepal. Continuous records are telemetered to
either the national seismic center in Kathmandu, or the regional seismic center in Birendranagar. The locations of the 3-component temporary
seismic stations operated between July and December 1995 are also shown.
 M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712 705

Fig. 3. Magnitude±frequency plots for all events detected between January 1996 and August 1998. Three zones with comparable areas are distin-
guished.

1833 earthquake (Bilham, 1995). Most of the others country. We ®rst describe the setting of the seismologi-
are probably related to smaller magnitude earthquakes cal network, we next present the main features of
that would have occurred close to the Kathmandu val- Himalayan structural geology. We ®nally discuss the
ley. No very large earthquake seems to have struck key features of Nepalese seismicity and their impli-
Far-Western and Western Nepal over the past few cen- cations for seismotectonics.
turies. This portion of the arc thus stands as a poten-
tial location for the next large Himalayan earthquake,
unless the seismotectonic behavior of that area is very 2. Characteristics of the seismological network
di€erent from that of the rest of the Himalayan arc.
In order to improve knowledge of the seismotec- A telemetered seismological network of 5 vertical
tonics of Nepal, the Department of Mines and short period stations was installed in 1985 around
Geology (DMG, HMG, Nepal), in collaboration with Kathmandu. This network has been expanded in 1994
the Laboratoire de DeÂtection et GeÂophysique (LDG, to a National Seismological Network consisting of 17
CEA, France) has been monitoring the seismic activity short period stations covering the entire east±west
in central Nepal since 1985, and over the whole stretch of the country (Fig. 2). The continuous records
country since 1994. Since earthquakes express brittle are telemetered to either the National Seismological
accommodation of deformation associated with short- Center in Kathmandu, or the regional seismological
ening and mountain building across the Himalayas, center in Birendranagar, where they are processed. The
this monitoring was completed by geological investi- velocity model used for the whole country is the same
gations in order to constrain deformation rates on the as for Central Nepal and consists of a three layered
long term (Lave and Avouac, 1999), and geodetic depth model with P-wave velocities of 5.6, 6.5 and
measurements of on-going interseismic deformation, in 8.1 km/s and interfaces at 23 and 55 km depth
collaboration with the LDG and CNRS (France) as (Pandey, 1985).
well as CIRES (Colorado, USA). From the magnitude±frequency distribution of the
In this paper, we report the results from two years earthquakes detected within Nepal, we estimate the
of monitoring of seismic activity over the whole detection threshold to be about 2.0 ML (Fig. 3). The
 706
M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712

Fig. 4. Microseismicity map of Nepal monitored between May 1994 and January 1998. The blue lines AA', and BB' show locations of cross sections in Figs. 5 and 6. Active faults modi®ed
from Nakata (1989) within Nepal and from Armijo et al. (1986) for Southern Tibet. MCT and MBT from Amatya and Gnawali (1994).
 M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712 707

threshold in Western Nepal seems lower than in Nepal from the analysis of the 1985±1995 data
Central or Eastern Nepal essentially because most of recorded by the network around Kathmandu (Pandey
seismicity occurs closer to the seismic stations (Fig. 4). et al., 1995). It is clear from Fig. 4 that this seismic
The network was temporarily augmented by three, belt extends throughout Nepal from west to east. It
three-component short period stations in 1995. These mainly consists of earthquakes with magnitude 2±4
stations were operated between 07/1995 and 12/1995 (Fig. 4). There are lateral variations of the geometry
north of Kathmandu so as to surround the area of and activity of that belt. It is relatively narrow and
particularly intense seismic activity described by straight for about 550 km between 82.58E and 878E.
Pandey et al. (1995) (Fig. 2). Due to the improved azi- East of 878E the seismic belt is o€set by 50 km to the
muthal coverage, and to the unambiguous picking of north, and continues farther east at least 150 km. It
S-wave arrival times on the horizontal components, gets more di€use and complex in the Far-West. Two
constraints on the location and depth of epicenters in parallel trends, about 60 km apart can be distinguished
this area were signi®cantly improved. between 81.58E and 82.58E. West of 81.58E seismicity
Epicenters computed exclusively from the permanent concentrates again along a single but relatively di€use
network or including the three additional stations do linear trend. Two major discontinuities of the seismic
not di€er systematically as to their locations. The var- belt are thus identi®ed at 828E and 878E which appear
iance of the di€erence of locations is 4 km on average to be associated with clusters of seismic activity.
and may be taken to represent the uncertainty at the The largest event in Central Nepal, with magnitude
1ÿs con®dence level on locations determined from the 5.9 ML, occurred 30 km to the northeast of
permanent network. The depths computed from the Kathmandu on January 31 1997, at a depth of about
permanent stations seem to be systematically too large 20 km. The focal mechanism is a reverse fault with a
by about 5 km on average with an additional 4 km slight component of right lateral strike slip on a
random uncertainty. This bias might be due to a N608E striking fault dipping 808SW (Kumar, in press).
slightly erroneous velocity model associated with the The main event in Far-Western Nepal is the 6.3 ML
fact that the seismicity is mostly located outside the that occurred on January 5 1997 at an hypocentral
permanent network. In the following, depth determi- depth of 4±5 km.
nations are not corrected for this possible bias. Hypocentral depths are reasonably well constrained
only for Central and Western Nepal. They are gener-
ally shallower than about 30 km and tend to cluster at
3. Some characteristics of Nepal seismicity a mid-crustal level at depths between 10 and 20 km
(Fig. 6).
3.1. General features In terms of seismic moment, seismicity along the lin-
ear belt in Central Nepal has released about 7.9 1017
The territory of Nepal is characterized by a very N.m/yr. (computed for a 207  66 km2 window strik-
intense microseismic activity. Some 11,000 local and ing 1088E between longitudes 84.58E and 86.58E). In
regional events have been recorded and processed over Eastern Nepal it has released 3.3 1017 N.m/yr. (com-
the two and half years of operation of the National puted for a 200  110 km2 window striking 958E
Network. All events with magnitude above 2.0 are between longitudes 86.58E and 88.58E). In Far-
reported on the map shown in Fig. 4. The whole area Western Nepal the released moment amounts to about
is seismically active, but signi®cant lateral variations 1.0 1017 N.m/yr. (computed for a 240  90 km2 win-
are observed. Microseismic activity is particularly dow striking 1188E between longitudes 80.58E and
intense in Eastern and Far-Western Nepal (Fig. 2). 82.58). Western Nepal is much more quiet with about
Some patches of seismicity are also detected in 4.2 1016 N.m/yr. (computed for a 210  75 km2 win-
Southern Tibet. The frequency±magnitude relation fol- dow striking 1088E between longitudes 82.58E and
lows the Gutenberg±Richter law. For magnitudes in 84.58E). All along the belt, the bulk of the seismic
the range 2.0 to 5.5 ML, it yields a b-value between moment is released at depths between 10 and 30 km.
0.75 and 0.95 (Fig. 3) that does not seem to vary sig-
ni®cantly from one area to the other (Fig. 3).
3.3. Seismicity beneath the Sub-Himalaya
3.2. Seismicity beneath the Higher and the Lesser
Himalaya The epicentral zone of the 08/20/1988 ML=6.5
Udayapur earthquake in Eastern Nepal, is the only
The major feature of microseismicity in Nepal is the seismically active zone in the Sub-Himalaya (Fig. 4).
narrow belt of seismicity that follows approximately On-going seismic activity occurs at sub-crustal depths,
the topographic front of the Higher Himalaya (Fig. 4). between 35 and 55 km, in agreement with the hypocen-
This characteristic was ®rst recognised in Central tral depths determined for the 1988 main event [50 km
708 M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712

Fig. 5. (a) Geological section across the central Himalaya of Nepal (AA' in Fig. 4). Modi®ed from Brunel (1986). (b) Geological section across
the Himalaya of Far-Western Nepal (BB' in Fig. 4) after DeCelles et al. (1998).

after Chen and Kao (1996)] and aftershocks (Pandey Himalaya consists mainly of foliated metasediments of
and Nicolas, 1989). probably Precambrian to Paleozoic ages. As shown in
section AA ' (Fig. 5), the Lesser Himalaya form a large
antiform, the Pokhara-Gurka anticlinorium of PeÃcher
4. Geological framework (1978), that can be traced from Central (Brunel, 1986;
PeÃcher, 1978), to Eastern (Schelling, 1992) and Far-
4.1. Structure of the Himalaya of Nepal Eastern Nepal (Schelling and Arita, 1991). In Far-
Western Nepal the tectonic style of the Lesser
The large-scale geological structure of the Himalaya suggests a wider duplex with many north-
Himalayan range shows little lateral variation over its dipping imbricated sheets (section BB ' in Fig. 5). A
2000 km long stretch (Gansser, 1964). For example, similar structure is proposed for the Lesser Himalaya
Fig. 5 shows structural sections across the Himalaya
of the Garhwal±Kumaon area (Srivastava and
of Central Nepal [modi®ed from Brunel (1986)] and
Gautam, 1994). The transition between the two styles
Far-Western Nepal [from DeCelles et al. (1998)]. Three
of tectonics in the Lesser Himalaya occurs at about
main domains are generally de®ned. The Higher
Himalaya consists of amphibolite-grade schist and 828E.
gneiss intruded by leuco-granitic plutons. These units Several crystalline thrust sheets were thrust on top
are bounded to the south by the Main Central Thrust of the Lesser Himalaya. They form generally pinched
(MCT) fault, a several kilometer wide fault zone that synformal klippens (Fig. 4 and 5). The Lesser
follows the range front of the Higher Himalaya (Le Himalaya is bounded to the south by the Main
Fort, 1975). South of the Higher Himalaya, the Lesser Boundary Thrust (e.g., Gansser, 1964). South of it, the
Himalaya forms the footwall of the MCT. The Lesser Sub-Himalaya consists of Tertiary molasse involved in
 M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712 709

thin-skinned tectonics induced by thrusting of the et al., 1992; Pandey et al., 1995; Bilham et al., 1997).
Himalaya over the Indian basement (Fig. 5). It has therefore been inferred that microseismicity
re¯ects stress accumulation during interseismic periods.
4.2. Active faults within Nepal and Southern Tibet Comparison with the structural section across Central
Nepal further indicates that this area coincides with
The known active faults around Nepal are shown in the ramp along the MHT inferred to connect the ¯at
Fig. 4. The major active fault along the Himalaya is decollement beneath the Lesser Himalaya with deep-
the Main Frontal Thrust that marks the southern edge seated deformation beneath the Higher Himalaya (Fig.
of the Himalayan foothills (Nakata, 1989). Recent 6). This was inferred from the data collected by the
motion along the MFT is evident from the defor- Kathmandu network between 1985 and 1993 along
mation of Quaternary sediments (e.g., Yeats and Lillie, section AA ' (Pandey et al., 1995). The seismicity
1991) or of abandoned ¯uvial terraces of Pleistocene recorded since the upgrading of the network yields the
to Holocene ages (Delcaillau, 1986). The analysis of same pattern. The epicenters are clustered around the
folded Holocene terraces along the Bagmati and location of the ramp as inferred from surface geology
Bakeya rivers, south of Kathmandu, implies a slip rate (Fig. 6), and they release seismic moment that is also
on the MFT of 21.5 2 2 mm/yr (Lave and Avouac, well clustered in that area. Although more di€use in
1999). Although it is not clear that the this fault did Far-Western Nepal, the belt lies in a similar structural
rupture the surface during the largest Himalayan position (Fig. 5). This correlation suggests that the
earthquakes, this slip rate suggests that nearly all the ramp beneath the Higher Himalaya behaves as an
crustal shortening across the Himalaya is accommo- asperity focusing stress and strain build up in interseis-
dated by localized fault slipping along the MFT and mic periods (Pandey et al., 1995).
its northward continuation beneath the Lesser This seismotectonic scheme can be further tested
Himalaya. This major active fault is plotted as a thick from recently published GPS data and morphotectonic
line in Fig. 5. Its northward continuation beneath the investigations in the Sub-Himalaya. The modeling of
Higher Himalaya coincides with the mid-crustal re¯ec- leveling and GPS data (Jackson and Bilham, 1994;
tor located by Hirn et al. (1984) and Zhao et al. Bilham et al., 1997) indicates that the ¯at and the
(1993). Following Zhao et al. (1993), we call it the ramp must be locked in the interseismic period. The
Main Himalayan Thrust fault, MHT. geodetic signal was found to be consistent with con-
Other active faults have been identi®ed in Nepal but tinuous creep at depth beneath the Higher Himalaya
they are probably of secondary importance as seismic at a rate of 21 2 3 mm/yr, close to the long-term slip
hazards. Evidence for recent motion along the MBT rate on the MFT of 21.5 2 2 mm/yr (Lave and
was found mainly in Western Nepal (Nakata, 1989; Avouac, 1999). This is in keeping with the hypothesis
Mugnier et al., 1994). Nakata also found evidence for that the ramp and the ¯at are only activated during
right-lateral strike-slip motion along NW±SE trending very large earthquakes and allow transfer of nearly all
faults such as the Darma-Bari Gad fault system in the interseismic deformation to the most frontal struc-
Western Nepal (Nakata, 1982, 1989) (DBG in Fig. 4). tures (Pandey et al., 1995; Lave and Avouac, 1999).
Surface faulting was also documented at some places This scenario is consistent with Seeber and
along the MCT (Nakata, 1982, 1989). Armbuster's (1981) inference that large Himalayan
Active tectonics in southern Tibet is characterized earthquakes activate sub-horizontal detachment faults
by normal faulting along N±S grabens (Armijo et al., that extend beneath the Sub- and Lesser Himalaya.
1986). Some of the grabens, such as the Thakkhola Such a scenario is further supported by the modeling
graben cut across the Higher Himalaya (Fig. 3). East± of the coseismic elevation changes near Dehra Dun
West extension in Southern Tibet is thought to accom- (Chander, 1988; Gahalaut and Chander, 1997).
modate the variation of the thrusting azimuth along
the arcuate shape of the Himalaya (Baranowski et al., 5.2. Segmentation of the Main Himalayan Thrust fault
1984; Armijo et al., 1986).
As already discussed, the microseismic belt at the
front of the Higher Himalaya displays a nearly con-
5. Seismotectonic interpretation tinuous and straight trend between 828E and 878E.
This stretch coincides with the portion of the Lesser
5.1. Relation between microseismic activity and Himalaya characterized by a large antiformal anticline
interseismic stress build up as depicted on section AA ' (Fig. 5). This observation
suggests that the ramp along section AA ' probably
The belt of intense microseismic activity at the front extends as a single and linear feature all along that
of the Higher Himalaya in Central Nepal coincides stretch. This stretch ends to the east at the longitude
with a zone of uplift revealed in leveling data (Jackson of the Pum Qu Graben (086.58E), where the azimuth
710 M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712

Fig. 6. Density distribution of epicenters along AA ' and BB' sections. In order to take into account uncertainties in seismic locations, the distri-
bution of epicenters was ®ltered using an Gaussian ®lter with s=4 km. Fault geometries reported from Fig. 5. All events within 30 km from the
section are taken into account.

of the range front abruptly changes from about 1008E 878E may be considered as a single 600 km long seg-
to 908E (Fig. 4). The 08/20/1988 sub-crustal earth- ment across Central and Western Nepal. The extent of
quake beneath the Sub-Himalaya also occurred in that damage during the 1934 Bihar±Nepal event suggests
area. That junction is thus probably a major transverse that only the eastern portion of that stretch was acti-
tectonic feature. vated during that particular earthquake (Fig. 1).
To the west, the linear and straight seismicity zone Actually, in spite of the geometric continuity of the
ends at about 828E. At this longitude the seismicity seismicity feature, we may distinguish two smaller seg-
map shows two parallel belts that may suggest the ments on the basis of the very di€erent seismic activity
existence of two ramps, although surface structure in in Central and Western Nepal. The divide would lie at
that area is not documented well enough to support about 858E near the longitude of the Gyirong Graben
such an hypothesis. This zone correlates with a reen- (Fig. 4).
trant of the MCT (Bashyal, 1986; Amatya and
Gnawali, 1994) (Fig. 3). This correlation suggests that 5.3. Seismicity in Southern Tibet
Himalayan structures have resulted from cumulative
deformation over several tens of million years in a way Although the network does not provide very accu-
that is similar to active deformation at present, or that rate locations for events in Southern Tibet (they may
the present tectonics is partly controlled by inherited be wrong by a few tens of kilometers for small events),
structures. Whichever explanation is preferred, the seg- epicenters tend to be localized near the main active N±
mentation inferred from the seismicity pattern may be S normal faults identi®ed in southern Tibet (Armijo et
long-lived. al., 1986) (Fig. 4). Over the last two years, there has
The portion of the Himalayan arc between 828E and been a particularly intense and continuous activity
 M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712 711

along the Pum Qu Graben. The seismicity along the coseismic slip. Assuming that the long term slip rate
Gyirong Graben is dominated by a crisis that started on the MFT of 21.5 2 2 mm/yr determined at the
in 1994. Continuous moderate seismicity has been longitude of Kathmandu does not vary signi®cantly
recorded along the Thakkhola and Kung Co Grabens. along the Himalaya of Nepal, each segment should
North of the Yarlung±Tsangpo, some sporadic clusters rupture every about 130±260 years during M > 8
of seismicity seem also to correlate with active N±S earthquakes. This reasoning probably overestimates
grabens cutting across the Gangdise belt (Fig. 3). the probability of these large earthquakes because it
However, depth determinations for events so far from neglects the amount of deformation accommodated by
the National Seismological Network of Nepal are possible aseismic slip and smaller more frequent earth-
poorly constrained and may be biased. quakes.

5.4. Seismicity in the upper mantle beneath the Sub-

Himalaya Acknowledgements

The 08/20/1988 ML=6.5 Udayapur earthquake, that The authors are most grateful to all the sta€ of the
occurred in the upper mantle beneath the Sub- National Seismological Centre (NSC), DMG and
Himalaya, has no equivalent along the entire arc LDG who have contributed to the installation and
(Chen and Kao, 1996). Similar deep events have been maintenance of the permanent and mobile stations and
recorded beneath the northern foothills of the Tien routine processing of the seismic data. G.R. Chitrakar
Shan (Nelson et al., 1987). These unusual events may and B. Ka¯e of NSC are acknowledged for editing the
be related to compressional bending stresses in the routine processing of seismic data. Figures were
upper mantle induced by ¯exure of the footwall at the drafted by F. Guillois and E. Thauvin. We bene®ted
front of the over-thrusting mountain range. Such an from reviews by N. Seeber and Anne Paul. K. Burke
interpretation implies a good mechanical coupling and Dr Upreti provided helpful comments for the
between the crust and the upper mantle for the devel- preparation of the ®nal version of the manuscript.
opment of a single neutral zone in the mid-crust
(Burov and Diament, 1996). The location of that event
at the longitude of the Pum Qu graben, where we infer References
a major junction between two segments of the
Himalayan arc, might therefore not be fortuitous. Amatya, K.M., Gnawali, B.N., 1994. Geological map of Nepal,
1:1000000 scale, Department of Mines and Geology, Kathmandu,
ICIMOD, CDG, UNEP.
6. Conclusion Armijo, R., Tapponnier, P., Mercier, J.L., Tonglin, H., 1986.
Quaternary extension in Southern Tibet. Journal of Geophysical
Research 91, 13,803±13,872.
The main feature of the seismicity of Nepal is a belt Baranowski, J., Armbuter, J., Seeber, L., Molnar, P., 1984. Focal
of seismicity following the front of the Higher depths and fault plane solutions of earthquakes and active tec-
Himalaya that probably re¯ects interseismic stress ac- tonics of the Himalaya. Journal of Geophysical Research 89,
6918±6928.
cumulation. Lateral variations along that belt may
Bashyal, R.P., 1986. Geology of Lesser Himalaya, far western Nepal.
relate to the segmentation of the Himalayan arc. This Sciences de la Terre, Memoire 47, 31±42, Nancy, November.
segmentation is probably a long-lived feature because Bilham, R., 1995. Location and magnitude of the 1833 Nepal earth-
it correlates with lateral variation of geological struc- quake and its relation to the rupture zones of contiguous great
ture. The seismicity suggests a division into four 250± Himalayan earthquakes. Current Science 69 (2).
300 km long segments with major junctions at about Bilham, R., Larson, K., Freymuller, J. Project IDYL-HIM members,
1997 Indo-Asian Convergence rates in the Nepal Himalaya,
828 and 878E and a less obvious one at about 858E. Nature 386, 61±66.
These junctions are aligned with some of the major Brunel, M., 1986. Ductile thrusting in the Himalayas: shear sense cri-
grabens of Southern Tibet, that are also associated teria and stretching lineations. Tectonics 5, 247±265.
with microseismic activity. Such segments may produce Burov, E., Diament, M., 1996. Isostasy, equivalent elastic thickness,
and inelastic rheology of continents and oceans. Geology 24,
characteristic earthquakes (Schwartz and Coppersmith,
419±422.
1984) similar to the 1905 Kangra or 1934 Bihar±Nepal Chander, R., 1988. Interpretation of observed ground level changes
earthquake. The coseismic slip during the 1934 Bihar± due to the 1905 Kangra earthquake, Northwest Himalaya.
Nepal earthquake is estimated to be of the order of Tectonophysics 149, 289±298.
4.7 m from the estimated seismic moment (Molnar and Chen, W.P., Kao, H., 1996. Seismotectonics of Asia: some recent
progress. In: Yin, A., Harrison, M. (Eds.), The tectonic of Asia.
Deng, 1984). For the 1905 Kangra earthquake, leveling
Cambridge University Press, pp. 37±62.
data in Dehra Dun are consistent with 3±5 m of Chitrakar, G.R., Pandey, M.R., 1986. Historical earthquakes of
coseismic slip (Chander, 1988). We may therefore Nepal. Bulletin of the Geological Society, Nepal 4, 7±8.
assume a characteristic earthquake with 3±5 m of DeCelles, P.G., Gehrels, G.E., Quade, J., Ojha, T.P., Kapp, P.A.,
712 M.R. Pandey et al. / Journal of Asian Earth Sciences 17 (1999) 703±712

Upreti, B.N., 1998. Neogene foreland basin deposits, erosional Nelson, M.R., McCa€rey, R., Molnar, P., 1987. Source parameters
unroo®ng, and the kinematic history of the Himalayan fold-thrust for 11 earthquakes in the Tien Shan, central Asia, determined by
belt, western Nepal. Geological Society of America Bulletin 119 P and SH waveform inversion. Journal of Geophysical Research
(1), 2±21. 92, 12,629±12,648.
Delcaillau, B., 1986. Dynamique et evolution morphostructurale du Pandey, M.R., 1985. Seismic model of Central and Eastern Lesser
piemont frontal de l'Himalaya: les Siwalik de l'Himalaya du Himalaya of Nepal. Journal of the Geological Society, Nepal 3,
Nepal oriental. Revue de GeÂographie Physique et GeÂologie 1±11.
Dynamique 27, 319±337. Pandey, M.R., Molnar, P., 1988. The distribution of intensity in the
DeMets, C., Gordon, R.G., Argus, D.F., Stein, S., 1990. Current Bihar-Nepal earthquake of 15 January 1934 and bounds of the
plate motion. Geophysical Journal International 101, 425±478. extent of the rupture. Journal of the Geological Society, Nepal 5,
Gansser, A., 1964. In: Geology of the Himalayas. Interscience 22±44.
Publisher, New York 289 pp. Pandey, M.R., Nicolas, M., 1989. The aftershock sequence of the
Gahalaut, V.K., Chander, R., 1997. Evidence for an earthquake Udayapur earthquake of August 20, 1988. Journal of the
cycle in the NW outer Himalaya near 788E longitude, from pre- Geological Society, Nepal.
cision levelling observation. Geophysical Research Letters 24, Pandey, M.R., Tandukar, R.P., Avouac, J.P., LaveÂ, J., Massot, J.P.,
225±228. 1995. Interseismic Strain Accumulation on the Himalayan Crustal
Hirn, A.M., Sapin, J.C., LeÂpine, Z.X., Xu, E.Y., Gao, J.W., Teng, Ramp (Nepal). Geophysical Research Letters 22, 751±754.
Pandey, M.R., 1984. Himalaya border of Tibet: crustal structure PeÃcher, A., 1978. DeÂformation et meÂtamorphisme associeÂs aÁ une
and variability. Nature 307, 23±25. grande zone de cisaillement, exemple du grand chevauchement
Jackson, M., Barrientos, S., Bilham, R., Kyestha, D., Buddhi, central Himalayen (MCT), transversales des Annapurnas et du
Shrestha, 1992. Uplift in the Nepal Himalaya revealed by spirit
Manaslu, NeÂpal. TheÁse d'Etat, Grenoble, 310 pp.
levelling. Geophysical Research Letters 19, 1539±1542.
Schelling, D., 1992. The tectonostratigraphy and structure of the
Jackson, M., Bilham, R., 1994. Constraints on Himalayan defor-
eastern Himalaya. Tectonics 11, 925±943.
mation inferred from vertical velocity ®elds in Nepal and Tibet.
Schelling, D., Arita, K., 1991. Thrust tectonics, crustal shortening
Journal of Geophysical Research 99, 13,897±13,912.
and the structure of the Far-Eastern Nepal Himalaya. Tectonics
Kumar, B., 1999. Aftershock distribution and focal mechanism of
10, 851±862.
Sarshing earthquake (1/31/1997), IISEEE, Seismology Course,
Seeber, L., Armbuster, J., 1981. Great detachement earthquakes
pp. 97±98, Tsukuba, Japan (in press).
along the Himalayan arc and the long term forecasts. In:
LaveÂ, J., Avouac, J.P., 1999. Active folding of abandoned Fluvial
Simpson, E.W., Richards, P.G. (Eds.), Earthquake Prediction:
Terraces across the Siwalik Hills (Nepal), Journal of Geophysical
Research (in press). An International Review, Maurice Ewing Series 4. American
Le Fort, P., 1975. Himalayas: the collided range, present knowledge Geophysical Union, Washington, DC, pp. 259±277.
of the continental arc. American Journal of Science 275A, 1±44. Schwartz, D.P., Coppersmith, K.J., 1984. Fault behavior and charac-
Middlemiss, C.S., 1910. The Kangra earthquake of 4th April 1905. teristic earthquakes: examples from the Wasatch and San
Memoirs of the Geological Survey, India 37, 1±409. Andreas fault zones. Journal of Geophysical Research 89, 5681±
Molnar, P., Deng, Q.D., 1984. Faulting associated with large earth- 5698.
quakes and the average rate of deformation in central and eastern Srivastava, Praveen, Gautam, Mitra, 1994. Thrust geometrics and
Asia. Journal of Geophysical Research 89, 6203±6228. deep structure of the outer and lesser Himalaya, Kumaon and
Molnar, P., Pandey, M.R., 1989. Rupture zones of great earthquakes Garhwal (India): Implications for evolution of the Himalayan
in the Himalayan region. In: Proc. Indian Acad. Sci. (Earth and fold-and-thrust belt. Tectonics 13 (1), 89±109.
Planetary Science), 98 (1), pp. 61±70. Yeats, Lillie, 1991. Contemporary tectonics of the Himalayan frontal
Mugnier, J.L., Huyghe, P., Chalaron, E., Mascle, G., 1994. Recent fault system: folds, blind thrusts and the 1905 Kangra earth-
movements along the Main Boundary Thrust of the Himalayas: quake. Journal of Structural Geology 13, 215±225.
normal faulting in an over-critical thrust wedge? . Tectonophysics Yeats, R.S., Nakata, T., Farah, A., Fort, M., Mirza, M.A., Pandey,
238, 199±215. M.R., Stein, R.S., 1992. The Himalayan Frontal Fault Sytem.
Nakata, T., 1982. A photogrametric study on active faults, in the Annales Tectonicae 6, 85±98 (Special Issue, Suppl.).
Nepal Himalayas. Journal of Nepal Geological Society 2, 67±80. Zhao, W., Nelson, K.D., Project, I.N.D.E.P.T.H. Team, 1993. Deep
Nakata, T., 1989. Active faults of the Himalaya of India and Nepal. seismic re¯ection evidence for continental underthrusting beneath
Special paper Geological Society of America 232, 243±264. southern Tibet. Nature 366, 557±559.