Journal of the Geological Society, London, Vol. 156, 1999, pp. 457–464. Printed in Great Britain
Timing of formation of forebergs in the northeastern Gobi Altai, Mongolia:
implications for estimating mountain uplift rates and earthquake recurrence intervals
LEWIS A. OWEN 1 , DICKSON CUNNINGHAM 2 , BENEDICT W. M. RICHARDS 3 , EDWARD
RHODES 4 , BRIAN F. WINDLEY 2 , DORJ DORJNAMJAA 5 & JALBUUGIN BADAMGARAV 5
1
Department of Earth Sciences, University of California, Riverside, CA 92521-0423, USA
(e-mail: [email protected])
2
Department of Geology, University of Leicester, Leicester LE1 7RH, UK
3
Department of Geography, Royal Holloway, University of London, Egham, Surrey TW20 0EX, UK
4
Research Laboratory for Archaeology and History of Art, Oxford University, 6 Keble Road, Oxford OX1 3QJ, UK
5
Mongolian Academy of Sciences, Institute of Geology, Ulaan Bator, Mongolia
Abstract: Bedrock and Quaternary alluvial fans in the forelands of the northern Gobi Altai have been
faulted and warped by foreland-propagating thrust faults to form elongate hills known as forebergs.
Forebergs near Bogd town, north of Arsta Bogd, and at Ih Hetsüü, north of Baga Bogd were mapped and
their formation constrained using luminescence dating of lacustrine and fanglomerates sediments to give
localized uplift rates of c. 0.05 m ka 1, and between c. 1.6 and 4.1 m ka 1, respectively. Comparisons
between vertical surface rupture during the 1957 Gobi Altai earthquake with the age and amount of uplift
of lacustrine sediments provide a tentative estimate of between 1288382 to 497120 years for the
average recurrence interval of great earthquakes in the Baga Bogd region. These localized uplifts rates
support the view that the Gobi Altai Mountains probably took between 50 and 25 million years to form
and were, therefore, coeval with the uplift of the Himalayas and the Tibetan Plateau.
Keywords: Gobi Altai, Mongolia, earthquakes, luminescence dating, uplift rates.
The surface ruptures of the 1957 Gobi Altai (sometimes spelt
Altay) earthquake (Magnitude 8.3) in Mongolia has attracted
much interest in recent years because of its similarities with the
simultaneous rupture along both the San Andreas fault and
the Sierra Madre-Cucamonga fault during the 1857 Fort Tejon
earthquake (Bayarsayhan et al. 1996; Kurushin et al. 1997).
The principal Gobi Altai ruptures stretch for 260 km along the
northern margin of the mountains of Ih Bogd and Baga Bogd
to Bayan Tsagaan Nuruu, and for over 70 km along the
southern margin of Ih Bogd. The ruptures are characterized by
both left-lateral strike-slip, and thrust and reverse fault offsets.
Full and comprehensive details are given in Baljinnyam et al.
(1993) and Kurushin et al. (1997). The 1957 quake is believed
to be a consequence of slip along the E–W-trending left-lateral
North Gobi Altai fault system which is accommodating some
of the intracontinental strain caused by the distant India–
Eurasia collision 2500 km to the south (Tapponnier & Molnar
1979; Cunningham et al. 1996). Detailed structural traverses in
the eastern Gobi Altai support the view that the Gobi Altai
formed by localized transpressional deformation and uplift
along left-lateral strike-slip faults and at stepover zones
between fault segments (Cunningham et al. 1997; Owen et al.
1999).
The Gobi Altai Mountains rise to a maximum of 3957 m
a.s.l. in Ih Bogd from an extensive desert surface that has
elevations of between 1500 and 2000 m a.s.l. Large alluvial
fans prograde from the mountain ranges. In the forelands of
the Gobi Altai Mountains, the alluvial fans and bedrock are
faulted to produce low hills that are approximately parallel to
the main mountain front. Florensov & Solonenko (1963)
referred to these low hills as ‘forebergs’ and considered them to
have been formed by thrust slip on faults that dip steeply
beneath the hills and continue at gentle angles towards the
neighbouring mountain fronts. These landforms were also
described in Owen et al. (1997, 1998) and Kurushin et al.
(1997), but little attention had been given to their evolution
Fig. 1. Topographic map of the
northeastern Gobi Altai showing the
locations of the forebergs discussed in
this paper. The central and eastern
stretch of the 1957 Gobi Altai
earthquake rupture zone is shown by the
thick black line (adapted from Florensov
& Solonenko 1963 and Baljinnyam et al.
1993).
457
MS
FS
FS
FS
FS
FS
TB01
TB09
TB09
TB10
TB11
TB11
Fl
L
L
L
L
L
4438.401N/10207.286E
4454N/10141E
4454N/10141E
4454N/10141E
4454N/10141E
4455N/10141E
Location
CGQ
FGP
CGQ
FGP
FGP
FGQ
Type‡
3.96
2.22
2.22
3.52
3.45
3.45
U sed.
(ppm)
11.7
9.28
9.28
14.5
14.6
14.6
Th sed.
(ppm)
2.01
2.79
2.79
2.59
2.63
2.63
K sed.
(%)
180
155
155
255
220
220
Cosmic
doserate
(ìGy a 1)
21
1.4
1.4
2.9
3.7
3.7
Water
content
(%)
3140240
5380430
4177331
6490580
6360590
5551541
Total
doserate
(ìGy a 1)
DE
(Gy)
99.817.5
145.622.8
121.127.6
92.911.5
104.413.6
63.215.6
DE is equivalent dose (in Gy).
Uncertainties on INAA analyses of U, Th and K are taken to be 10% with the detection limits being 0.05% for K, 0.1 ppm for Th and 0.5 ppm for U.
*Lithology: MS, medium sand; FS, fine sand.
†Environment of deposition: L, lacustrine sediment; Fl, fluvial sediment.
‡Type: FGP, fine-grained (4–11 ìm) polymineralic (combined IRSL and OSL results); FGQ, fine-grained quartz OSL; CGQ, coarse grained (90–125 ìm) quartz OSL.
Lithology*
Sample
Environment
of deposition†
Table 1. Values used to calculate luminescence ages and results
31.76.1
27.04.8
29.07.0
14.32.2
16.42.6
11.43.0
Age
(ka)
458
LEWIS A. OWEN ET AL.
TIMING OF FOREBERG FORMATION IN THE NE GOBI ALTAI, MONGOLIA
459
Fig. 2. Views of the easternmost end of the Bogd foreberg looking SE towards the north side of Artsa Bogd. The foreberg forms an inlier
within the Quaternary alluvial fans.
and rates of formation. Owen et al. (1997) provided tentative
rates of vertical deformation of alluvial fan surfaces of between
approximately 0.1 and 1 m ka 1. They based these calculations on tentative relative ages of the alluvial fan deposits
which they believed were deposited during the global Last
Glacial Maximum.
In this paper we examine the structure and geomorphology
of forebergs in two areas and date Quaternary lacustrine and
fanglomerates sediments to constrain the age of foreberg
formation. This age control is used to calculate localized uplift
rates and to estimate recurrence intervals for great earthquakes
in the north Gobi Altai.
prise south dipping Cenozoic and Cretaceous limestones, conglomerates, sandstones and marlstones, deformed by steeply
dipping, north-verging thrust faults. The forebergs are
bounded to the north by a north-verging upright fold (Fig. 3b).
Quaternary alluvial fans are deposited against the bedrock
hills. At the western end of the easternmost hill, Quaternary
fanglomerates are upwarped and cut by steep thrust faults
(Fig. 3c & d). An alluvial fan sediment sample was collected
for luminescence dating and its location is shown in Fig. 3d.
Measurements of fault offsets at this location and simple
leveling across the deformed alluvial fan surface show at least
1.5 m vertical displacement.
Geomorphology and structure
Ih Hetsüü forebergs
Two main study areas, the Bogd and Ih Hetsüü forebergs, were
mapped with the aid of Landsat MSS images and using a
hand-held global positioning receiver (Fig. 1). Sedimentary
sections were measured in the forebergs and in the alluvial
fans, and lacustrine and alluvial fan sediment samples were
collected for luminescence dating (Table 1).
Bogd forebergs
The Bogd forebergs comprise three elongate hills that occur
c. 10 km SW of Bogd town and 4 km north of Arsta Bogd
(Figs 2 & 3). These hills trend NE, parallel to lineaments visible
on Landsat images, and the hills rise to a maximum elevation
of 120 m above the surrounding alluvial fans. The hills com-
The Ih Hetsüü forebergs are located 4 km north of Baga Bogd
(Figs 1, 4 & 5). The youngest example, the Hetsüü foreberg, is
adjacent to the active trace of the 1957 Gobi-Altai earthquake
rupture and is described in detail by Kurushin et al. (1997). To
the east of the Hetsüü foreberg, a series of elongate hills rise to
130 m above the recent alluvial fans (Fig. 4a, 5 & 6). Each hill
is asymmetric with a steep northern face. The hills consist of
bouldery and cobbly fanglomerates that dip southward at
angles between 8 and 12 (Fig. 4b). Lacustrine deposits are
present south of several of these ridges and were formed when
drainage was blocked during the uplift of the forebergs. The
southernmost lacustrine deposit extends for at least 600 m in a
north-south direction and is tilted southwards 41, while
the northernmost lacustrine deposit is folded into an open fold
460
LEWIS A. OWEN ET AL.
Fig. 3. Details of the Bogd foreberg. (a) Topographic map of part of the northern edge of Artsa Bogd showing the location of the Bogd
foreberg in relation to the main mountain bounding and foreland thrust faults. The location of this map is shown in Fig. 1. (b) Simplified
geological map and section through the part of the Bogd foreberg (adapted from Cunningham et al. (1997). (c) Geomorphic map showing the
typical topography and the location of deformed Quaternary sediments in part of the Bogd foreberg area. (d) Faulted marlstones and
Quaternary sediments exposed in a channel cutting through part of the Bogd foreberg (4438.40N/10207.29E) showing the location of
luminescence dating sample TB01 and lower hemisphere stereoplot of the fault plane orientations. The location of the section is shown in (c).
with the southern limb dipping at 71 that extends for at
least 550 m (Fig. 6a & b). Assuming the lacustrine beds were
originally deposited horizontally, the southernmost deposit
has, therefore, been uplifted by c. 4210 m (600 m
Sin41) and northernmost deposit by c. 6712 m
(550 mSin71). The hills and lacustrine deposits are
deeply incised and provide good exposures for sampling.
The locations and stratigraphic position of each luminescence dating sample is shown in Fig. 6b. These deposits
and landforms are in turn overlain by recent alluvial fans
deposits.
Luminescence dating
Sediment samples from lacustrine and alluvial fans deposits were
collected in opaque plastic tubes for luminescence dating. These tubes
remained sealed until opened under controlled laboratory lighting.
Modern water content was determined by mass loss following drying
at about 50C for 24 hrs. Modern values are assumed to be typical of
water content throughout burial period, although there is no means of
verifying this assumption. Coarse (90–125 ìm) quartz and fine polymineral grains (4–11 ìm) fractions were prepared in the laboratory
using the methods described in Rhodes (1988), Zimmerman (1967) and
Rees-Jones (1995).
All luminescence measurements were made using an automated
Risø reader (TL-DA-12) fitted with infrared diodes emitting at
88080 nm used for infra red stimulated luminescence (IRSL)
measurements, and a filtered halogen lamp which provides wavelengths between 420 and 560 nm (2.9–2.2 eV) for green light optically
stimulated luminescence (OSL) measurements. Luminescence signal
emissions were filtered with two U340 and one BG39 glass filters.
Equivalent dose (DE) values were determined using a naturally normalized total integral, multiple aliquot additive dose technique, fitting
a single saturating exponential function (Grün & Brumby 1994).
Forty-eight aliquots were measured at eight dose points, with the
maximum additional beta dose being 4–5 times the preliminary DE
estimate. Subtraction of background was by the last integral subtraction method (Aitken & Xie 1992). The pre-heat treatment used for all
quartz sub-samples was 220C for 5 minutes and for polymineral
samples was 5 days at 100C followed by 4 hours at 160C. Measurement of all sub-samples was performed for 50 s of stimulation at room
temperature, for both IRSL and OSL. In the case of polymineralic
samples, IRSL measurement preceded OSL measurement using the
same aliquots. The magnitude of a thermal transfer component in the
quartz OSL signal (Rhodes & Bailey 1997) was determined using a
regenerative x-axis intercept. In all cases, the magnitude of the thermal
transfer signal was found to be insignificant. IRSL and OSL fading
tests were carried out for most of the samples measured, and showed
no discernible fading over a 10 day period. Longer term anomalous
fading cannot be ruled out for the polymineral sub-samples. Where
multiple DE values were measured for the fine-grained samples (both
quartz and polymineralic fractions), these were combined to produce a
weighted mean DE value.
TIMING OF FOREBERG FORMATION IN THE NE GOBI ALTAI, MONGOLIA
Fig. 4. Views of the Ih Hetsüü forebergs. (a) View looking SE towards the northern edge of Baga Bogd with the Ih Hetsüü forebergs in the
mid-ground. (b) View looking eastward at one of the Ih Hetsüü forebergs showing northerly dipping fanglomerates.
461
462
LEWIS A. OWEN ET AL.
Fig. 5. Landsat MSS image of the
Hetsüü and Ih Hetsüü forebergs in
relation to the northern edge of Baga
Bogd. The surface rupture produced
during the 1957 Gobi Altai earthquake is
shown by the dashed white line
(h=vertical offset in metres; half arrows
show horizontal offsets in metres).
Adapted from Kurushin et al. (1997).
The environmental dose rate was calculated from Neutron
Activation Analysis of U, Th and K content of each sample and the
water content measured for each sample was used to calculate
environmental radiation attenuation during burial. The cosmic dose
rate was calculated according to Prescott & Hutton (1994).
Table 1 shows the values used to determine sample ages and the
derived age estimates. DE values derived from IRSL and OSL of
polymineralic samples show no systematic differences and have been
combined. Sample TB09 is used to compare coarse-grained quartz
OSL and the fine grained polymineralic combined OSL and IRSL
results. There is no significant difference within the measurement
uncertainties. Similarly, sample TB11 shows that the fine-grained
quartz OSL date and the fine-grained polymineral combined OSL and
IRSL ages are coherent and consistent. These data lead to an increased
degree of confidence in the age estimates, both from the point of view
of optical resetting prior to burial and for thermal stability of the
luminescence signals.
Discussion and implications
The Bogd and Ih Hetsüü forebergs were formed by steep thrust
faults that propagate into the foreland from the main Gobi
Altai ranges. On the basis of the deformed lacustrine deposits
in the Ih Hetsüü forebergs, localized uplift rates are
1.60.6 m ka 1 and 4.11.0 m ka 1 (i.e., total uplift of
lacustrine deposits/age of lacustrine deposits=4210 m/
27.04.8 ka and 6712 m/16.42.6 ka) for the southernmost and northernmost forebergs, respectively. Given a
constant uplift rate of between about 1.60.6 to 4.11.0 m
ka 1 for the Ih Hetsüü forebergs, the highest foreberg may
have formed within c. 130 000 and 25 490 years (height of
foreberg/uplift rate=130 m/1.60.6 m ka 1 and 130 m/
4.11.0 m ka 1). Furthermore, by comparing the maximum
surface displacement (2 m) at the Hetsüü foreberg during the
1957 earthquake and the uplift of the lacustrine sediments at
Ih Hetsüü, i.e., 4210 m since 27.04.8 ka (TB09) and
6712 m since 16.42.6 ka (TB11), this gives an estimate of
c. 215 (4210 m/2 m) and 336 (6712 m/2 m) possible
great earthquakes. These calculations assume that all the uplift
was coseismic and only occurred during great earthquakes.
They also assume that 2 m of uplift is characteristic of all great
earthquakes at this location. Furthermore, Marco et al. (1996)
for example, showed that seismic activity is not necessarily
uniformly distributed through time rather it may be clustered
between periods of quiescence. Given these caveats, however,
it is possible to postulated that the recurrence interval for a
great earthquake in this region may be between 1288382
and 497120 years (27.04.8 ka/215 earthquakes and
16.42.6 ka/336 earthquakes).
Dating deformed fanglomerates across the Bogd foreberg
gave much lower localized minimum surface uplift rates
(0.050.003 m ka 1 =the amount of vertical offset of
fanglomerates/the age of the fanglomerates (TB01), i.e.,
1.50.1 m/31.76.1 ka) than for Ih Hetsüü. This suggests
that Arsta Bogd is less active than the Baga Bogd range and
this is consistent with the lower topography and more sinuous
mountain front of Artsa Bogd. Nevertheless, the whole of the
Bogd foreberg probably formed in less than 2.4 million years
(the maximum relative relief of the Bogd foreberg/vertical
uplift rate, i.e., 120 m/0.05 m ka 1).
Given that the localized minimum uplift rates are commonly
in the order of about 0.05 and 1 m ka 1 and that the relative
relief between the mountains and their forelands is between
2500 and 2000 m, it follows that the Gobi Altai was probably
uplifted over a period of between about 50 and 25 million
years. Such a conclusion, however, must be treated with care
because of the uncertainties associated with extrapolating
localized deformation rates to regional scales, and across time
scales which range from tens of thousands to millions of years.
Nevertheless, if these extrapolations are valid, the uplift of the
Gobi Altai is considerably longer than the very rapid uplift (<1
million years) proposed by Baljinnyam et al. (1993). These
conclusions help support the view that the uplift of the Gobi
Altai was coeval with the formation of the Himalayas and is
the result of the India–Eurasia collision. Furthermore, the low
rates of vertical deformation in this study are consistent with
the low rates (c. 1.2 m ka 1) of horizontal displacement
determined by Ritz et al. (1995) along the Bogd fault at the
eastern end of Ih Bogd. Conversely, it could be argued that the
low deformation rates determined in this study, and that of
Ritz et al. (1995), do not represent steady state deformation,
but rather the Bogd foreberg and many other active zones may
have been active for shorter periods at higher rates before
deformation ceased locally.
Constraining the style and deformation of forebergs and
alluvial fans in the forelands of the Gobi Altai is clearly
important for developing models and quantifying rates of
uplift in transpressional settings. Forebergs may also provide useful analogues for understanding the nature of fault
Fig. 6. (a) Geomorphic map of the Ih Hetsüü forebergs. Adapted
from Owen et al. (1997). See Fig. 5 for the location of the map.
(b) Geological section through part of the Ih Hetsüü forebergs and
graphic sedimentary logs showing the locations of the luminescence
dating samples TB09, TB10 and TB11.
464
LEWIS A. OWEN ET AL.
propagation and active folding in other major transpressional
fault systems.
This research was undertaken as part of NERC grant GR9/01881
awarded to BFW and LAO and NERC studentship GT4/95/199/E
awarded to BWMR. Thanks goes to British Petroleum for supplying
the Landsat MSS images.
References
A, M.J. & X, J. 1992. Optical dating using infrared diodes: young
samples. Quaternary Science Reviews, 11, 147–152.
B, I., B, A., B, B.A., C, A., D’,
M.G., G, L., K, V.M., K, R.A., M, P.,
P, H., V, Y.Y. 1993. Ruptures of major earthquakes and
active deformation in mongolia and its surroundings. Geological Society of
America, Memoir 181.
B, C., B, A., E, B., H, K.W.,
K, R.A., M, P. & Ö, M. 1996. 1957 Gobi-Altay,
Mongolia earthquake as a prototype for southern California’s most
devastating earthquake. Geology, 24, 579–582.
C, W.D., W, B.F., D, D., B, J. &
S, M. 1996. Late Cenozoic transpression in southwestern Mongolia
and the Gobi-Altai-Tien Shan Connection. Earth and Planetary Science
Letters, 140, 67–82.
——, ——, O, L.A., B, T., D, D. & B, J. 1997.
Geometry and style of partitioned deformation within a Late Cenozoic
transpressional zone in the eastern Gobi Altai Mountains, Mongolia.
Tectonophysics, 277, 285–306.
F, N.A. & S, V.P. (eds) 1963. The Gobi-Altai earthquake.
Moscow, Akademiya Nauk USSR [in Russian] [English translation by
Israel Program for Scientific Translations] 1965, Department of Commerce
Washington, DC, US.
G̈, R. & B, S. 1994. The assessment of errors in past radiation doses
extrapolated from ESR/TL dose-response data. Radiation Measurements,
23, 307–315.
K, R.A., B, A., L, M., E, B., M, P.,
B, C., H, K.W. & L, J. 1997. The surface rupture of
the 1957 Gobi-Altay, Mongolia, earthquake. Geological Society of America,
Special Paper, 320.
M, S., S, M. & A, A. 1996. Long-term earthquake clustering: a
50,000-year paleoseismic record in the Dead Sea Graben. Journal of
Geophysical Research, 11, B3, 6179–6191.
O, L.A., C, D., W, B.F., B, J. &
D, J. 1999. The landscape evolution of Nemegt Uul: a late
Cenozoic transpressional uplift in the Gobi Altai, southern Mongolia. In:
S, B. & W, B (eds) Uplift, Erosion and Stability: Geological
Society Special Publications, in press.
——, R, B., R, E.J., C, W.D., W, B.F.,
B, J. & D, D. 1998. Relic permafrost structures in
the Gobi of Mongolia: age and significance. Journal of Quaternary Science,
13, 539–548.
——, W, B.F., C, W.D., B, G. & D,
D. 1997. Quaternary alluvial fans in the Gobi Desert, southern Mongolia:
evidence for neotectonics and climate change. Journal of Quaternary
Science, 12, 239–252.
P, J.R. & H, J.T. 1994. Cosmic ray contributions to dose rates for
luminescence and ESR dating: large depths and long-term time variations.
Radiation Measurements, 23, 497–500.
R-J, J. 1995. Optical Dating of young sediments using fine-grain quartz.
Ancient TL, 13, 2, 9–14.
R, E.J. 1988. Methodological considerations in the optical dating of
quartz. Quaternary Science Reviews, 7, 395–400.
—— & B, R.M. 1997. The effect of thermal transfer on the zeroing of
luminescence of quartz from recent glacio-fluvial sediments. Quaternary
Geochronology (Quaternary Science Reviews), 16, 291–298.
R, J.F., B, E.T., B, D.L., P, H., S, A., R,
G.M., Y, F. & E, B. 1995. Slip rates along active faults
estimated with cosmic-ray-exposure dates: application to the Bogd fault,
Gobi-Altai, Mongolia. Geology, 23, 1019–1022.
T, P. & M, P. 1976. Slip-line theory and large-scale continental
tectonics. Nature, 264, 319–324.
Z, D.W. 1967. Thermoluminescence from fine grains from ancient
pottery. Archaeometry, 10, 26–28.
Received 19 May 1998; revised typescript accepted 5 October 1998.
Scientific editing by Alex Maltman.