Extensive glaciation in Transbaikalia, Siberia, at the Last Glacial
Maximum
Martin Margold a, *
, John D. Jansen a, 1
, Artem L. Gurinov b
, Alexandru T. Codilean c
,
David Fink d
, Frank Preusser a, 2
, Natalya V. Reznichenko e
, Charles Mifsud d
a
Department of Physical Geography, Stockholm University, SE-106 91, Stockholm, Sweden
b
Department of Geomorphology & Palaeogeography, Lomonosov Moscow State University, Leninskie gory, GSP-1, Moscow, 119991, Russian Federation
c
School of Earth & Environmental Sciences, University of Wollongong, Northfields Avenue, Wollongong, NSW, 2522, Australia
d
Institute for Environmental Research, Australian Nuclear Science & Technology Organisation, Locked Bag 2001, Kirrawee DC, NSW, 2232, Australia
e
Department of Geography, Durham University, Lower Mountjoy, South Road, Durham, DH1 3LE, UK
a r t i c l e i n f o
Article history:
Received 27 March 2015
Received in revised form
23 November 2015
Accepted 26 November 2015
Available online 14 December 2015
Keywords:
Glaciation
Transbaikalia
Last Glacial Maximum
Cosmogenic 10
Be exposure dating
Optically stimulated luminescence
a b s t r a c t
Successively smaller glacial extents have been proposed for continental Eurasia during the stadials of the
last glacial period leading up to the Last Glacial Maximum (LGM). At the same time the large mountainous
region east of Lake Baikal, Transbaikalia, has remained unexplored in terms of glacial chronology
despite clear geomorphological evidence of substantial past glaciations. We have applied cosmogenic
10
Be exposure dating and optically stimulated luminescence to establish the first quantitative glacial
chronology for this region. Based on eighteen exposure ages from five moraine complexes, we propose
that large mountain ice fields existed in the Kodar and Udokan mountains during Oxygen Isotope Stage 2,
commensurate with the global LGM. These ice fields fed valley glaciers (>100 km in length) reaching
down to the Chara Depression between the Kodar and Udokan mountains and to the valley of the Vitim
River northwest of the Kodar Mountains. Two of the investigated moraines date to the Late Glacial, but
indications of incomplete exposure among some of the sampled boulders obscure the specific details of
the post-LGM glacial history. In addition to the LGM ice fields in the highest mountains of Transbaikalia,
we report geomorphological evidence of a much more extensive, ice-cap type glaciation at a time that is
yet to be firmly resolved.
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
The glacial history of northern Eurasia has displayed opposing
trends between its western, central and eastern parts during the
last glacial cycle. Whereas the Fennoscandian Ice Sheet reached
close to its maximum Pleistocene extent at the Last Glacial
Maximum (LGM; in Oxygen Isotope Stage [OIS] 2), the continental
sector of the Barents-Kara Ice Sheet and smaller Eurasian ice
masses to the east show a sequence of successively diminishing
glacial advances (stadials) during the course of the last glacial cycle
(Svendsen et al., 2004; Stauch and Gualtieri, 2008; Astakhov, 2013).
In order to better understand this apparent anti-phase behaviour
between the Fennoscandian and Barents-Kara ice sheets, we focus
on major glaciated massifs in a remote area of Eurasia for which
information is wholly lacking.
Extensive mountain glaciation is documented for southern
Siberia, north-western Mongolia, and north-western China
(Lehmkuhl et al., 2011) where valley glacier advances are reported
for OIS-6 and -2 via optically stimulated luminescence and
cosmogenic nuclide exposure dating (Lehmkuhl et al., 2004, 2007;
Gillespie et al., 2008; Arzhannikov et al., 2012; Zhao et al., 2013).
Despite some regional differences in the relative prominence of
glacial advances during different cold periods, the general trend in
the Altai and Sayan mountains, located south of the maximum
continental limit of the Barents-Kara Ice Sheet, is for progressively
* Corresponding author.
E-mail addresses: martin.margold@natgeo.su.se (M. Margold), john.jansen@unipotsdam.de
(J.D. Jansen), gurinov.artem@gmail.com (A.L. Gurinov), codilean@uow.
edu.au (A.T. Codilean), david.fink@ansto.gov.au (D. Fink), frank.preusser@geologie.
uni-freiburg.de (F. Preusser), natalya.reznichenko@durham.ac.uk
(N.V. Reznichenko), charles.mifsud@ansto.gov.au (C. Mifsud).
1
Present address: Institute of Earth & Environmental Science, University of
Potsdam, Karl-Liebknecht-Str. 24-25, 14476 Potsdam-Golm, Germany.
2
Present address: Institute of Earth and Environmental Sciences, Albert-Ludwigs-Universit€at
Freiburg, Albertstr. 23-B, 79104 Freiburg, Germany.
Contents lists available at ScienceDirect
Quaternary Science Reviews
journal homepage: www.elsevier.com/locate/quascirev
http://dx.doi.org/10.1016/j.quascirev.2015.11.018
0277-3791/© 2015 Elsevier Ltd. All rights reserved.
Quaternary Science Reviews 132 (2016) 161e174
smaller glaciations since OIS-6 (Fig. 1; Gillespie et al., 2008;
Lehmkuhl et al., 2011; Arzhannikov et al., 2012; Zhao et al., 2013).
A large ice cap that existed in the central Sayan-Tuva Upland in OIS-
6 and -4 was apparently absent during OIS-3 and -2 when the
character of glaciation changed to branching valley glaciers
(Komatsu et al., 2007b; Arzhannikov et al., 2012). In the Khangai
Mountains of west-central Mongolia, the most extensive preserved
moraine ridges date to OIS-3 with OIS-2 ice limits reaching nearly
the same extents (Rother et al., 2014). An even stronger trend of
successively diminishing glacial advances is documented for the
Verkhoyansk Mountains, a large mountain group east of the Aldan
and Lena rivers in eastern Siberia (Fig. 1). Here moraine complexes
dated with infrared optically stimulated luminescence indicate
minimal ice masses during the (global) LGM (Stauch et al., 2007;
Stauch and Gualtieri, 2008; Stauch and Lehmkuhl, 2010).
Transbaikalia, situated between the Sayan and Verkhoyansk
mountains, is a vast mountainous region east of Lake Baikal where
intermontane basins of the Baikal Rift Zone are surrounded by
massifs with alpine relief (Figs. 1 and 2; Shahgedanova et al., 2002).
The region displays significant glacial modification but little information
is available concerning ice extent or chronology
(Shahgedanova et al., 2002; Margold and Jansson, 2011). Here we
investigate the timing of glaciations in central Transbaikalia with
particular focus on the Kodar Mountains, the region's highest
massif. Based on cosmogenic 10
Be exposure dating and optically
stimulated luminescence (hereafter abbreviated to exposure dating
and OSL, respectively), we propose a framework for the extent and
timing of the last glaciation and thus provide a new perspective on
the nature of the LGM in northern Eurasia.
2. Transbaikalia, Siberia
The mountains of Transbaikalia, including those surrounding
Lake Baikal, form a region of high-relief stretching more than
1000 km along a west-east axis between 108 and 121 E (Fig.1). The
mountain crux of the region is located between 58 and 55 N where
peaks ~3000 m above sea level (asl) stand more than 2000 m above
the surrounding extensional depressions of the Baikal Rift (Figs. 2
and 3). Lower mountains (<2000 m asl) are bounded to the north
by the River Lena at about 58 N, and south of 55e56 N sees the
rolling terrain of the Vitim Plateau. The regional climate is boreal,
with continentality and precipitation following a northwestsoutheast
(drying) gradient consistent with being at the easternmost
fringe of west-derived moisture (Geniatulin, 2000).
2.1. Glacial history
Evidence of past glaciation was first recognised in Transbaikalia
by Kropotkin (1876) who described traces of past glaciations in the
mountains of northern Transbaikalia down to 630 m asl. Describing
the mountains of the region, Obruchev (1923, 1932) distinguished
two glacial events and suggested that “glaciers had an alpine
character in more mountainous areas, while in plateau and highland
areas they were close to Scandinavian type” (Obruchev, 1923,
p. 34). Despite more intensive field research after 1945 motivated
by mineral exploration and infrastructure projects, no consistent
Komatsu et al. (2007a)Komatsu et al. (2007a)Komatsu et al. (2007a)
Fig. 1. Study area location within northern Eurasia plus information on the glacial history of the wider region. Location of Figs. 2 and 3 is indicated by a black rectangle (Mangerud et
al., 2004).
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174162
picture was to emerge regarding the glacial history of the region.
Opinions varied widely both in terms of the intensity of glaciation
and its timing. Logatchev (1974) suggested two glacial stages:
Samarovo (Middle Pleistocene) of valley glacier or ice-cap type, and
Zyryanka (Late Pleistocene), which included several stages of valley
glaciation. In contrast, Zolotarev (1974) suggested two stages of
valley glaciers during the middle Pleistocene, with the possibility
that some glacial sediments were of Early Pleistocene age, followed
by a more intensive glaciation in the early-Late Pleistocene when an
ice cap might have been present in the Bodaybo region (Figs. 2 and
3). The timing of this latter stage was determined by fossil fragments.
Climate during the LGM (Sartan stadial according to the
regional chronology) was too arid for the development of larger ice
masses, according to Zolotarev (1974), but valley glaciers existed in
the highest ranges.
More recent investigations of the sedimentary record within the
intermontane basins seem to confirm two glacial stages in the
Middle Pleistocene plus two stages during the Late Pleistocene
(Bazarov et al., 1981; Bazarov, 1986; Enikeev, 2008, 2009). In
addition, the intermontane basins Muya-Kuanda and Chara (Figs. 2
and 3) are described as having been repeatedly filled by lakes and,
although their origin remains disputed (cf. Kulchitskiy et al., 1995;
Ufimtsev et al., 1998; Kolomiets, 2008), many agree that the
damming of rivers by glaciers was responsible (Osadchiy, 1979,
1982; Krivonogov and Takahara, 2003; Enikeev, 2008, 2009;
Margold and Jansson, 2011; Margold et al., 2011)das suggested
by the pioneering work of Obruchev (1929).
None of the studies above provide chronological constraints on
the regional Quaternary glacial stratigraphy. Quantitative dating
available for the study area consists solely of 14
C and is summarised
below (all 14
C ages are calibrated using IntCal13 [Reimer et al., 2013]
calibration curve in Calib 7.0 online calibration calculator [Stuiver
and Reimer, 1993]; uncertainties are reported at 1s, with ages
rounded to the nearest 10 cal yrs; original uncalibrated 14
C ages and
laboratory numbers are given in parentheses) 3
.
1) Three samples are available for the Muya-Kuanda depression:
44,820e43,270 cal yrs BP (40,500 ± 930 14
C yrs BP; SOAN-2893),
and 42,880e41,860 cal yrs BP (38,320 ± 755 14
C yrs BP; SOAN-
2823) from a palaeosol in the bank of the Muya River
(Krivonogov and Takahara, 2003), and 43,360e37,350 cal yrs BP
(36,480 ± 3130 14
C yrs BP; SOAN-2483) on a piece of wood from
the vicinity of the Mudirikan and Muya river confluence
(Kulchitskiy et al., 1990; see Fig. 2 for the locations).
2) One sample, dated to 29,950e27,960 cal yrs BP (24,730 ± 770 14
C
yrs BP; SOAN-2979) from a piece of wood underlying lacustrine
sediments at the mouth of the Bambuyka river in the Vitim River
catchment upstream of the Muya-Kuanda Depression (Enikeev,
2009; see our Fig. 2).
3) A 1180 m deep drill-core in the Chara Depression, thought to
span the Pliocene and Pleistocene, is constrained by one date of
42,960e41,750 cal yrs BP (38,210 ± 870 14
C yrs BP; LU-977) on a
piece of wood, identifying OIS-3 sediments at 41.1e73 m depth
(Enikeev, 2008).
4) A large number of radiocarbon dates are available from several
sites in the northwest portion of the study area, largely from
Fig. 2. Study area. Locations of Figs. 4 and 7 are indicated by black rectangles; sites of radiocarbon dates discussed in the text are marked by stars.
3
We note that several of the reported 14
C ages approximate the ~40e35 kyr limit
of conventional 14
C dating and therefore possibly reflect a minimum age.
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174 163
archaeological research locales (Ineshin and Teten'kin, 2010).
Those relevant for the regional glacial history are: (i) a pair of
dates from the valley of the Karenga River (a tributary of the
Mama River; Fig. 2) on bones embedded in “moraine-like”
sediments dated to 20,990e20,700 cal yrs BP (17,290 ± 100 14
C
yrs BP; GIN-8983) and 29,370e27,900 cal yrs BP (24,600 ± 730
14
C yrs BP; SOAN-4422), with several other dates falling into a
broader LGM period derived from bones and mammoth tusks
embedded in laminated lacustrine sediments in the adjacent
tributary valleys of the Mama River; (ii) a date on charcoal from
an exposed section at the Vitim River, 1 km south of the
Mamakan River mouth, dated to 23,270e21,850 cal yrs BP
(18,670 ± 600 14
C yrs BP; SОАN-4546) and a date on bone from
an exposed section at the Bodaybo River 2 km upstream from its
mouth, giving 27,090e26,090 cal yrs BP (22,300 ± 500 14
C yrs
BP; GIN-8878; Fig. 2); (iii) a group of ages on bones and wood
from alluvial sediments on the Vacha River, in the area northeast
of Bodaybo (Fig. 2) spanning ~35,240 cal yrs BP (~33,000 14
C yrs
BP) to beyond the 14
C limit (Ineshin and Teten'kin, 2010).
The lack of either geochronology or systematic mapping of the
glacial geomorphology has kept Transbaikalia as an empty space on
the map between the more closely studied regions of the south
Siberian mountains (e.g., Lehmkuhl et al., 2004; Komatsu et al.,
2007a, b; Lehmkuhl et al., 2007; Gillespie et al., 2008; Lehmkuhl
et al., 2011; Arzhannikov et al., 2012; Zhao et al., 2013) and the
mountains of far north-east Russia and the Kamchatka peninsula
(e.g., Brigham-Grette, 2001; Bigg et al., 2008; Stauch et al., 2007;
Stauch and Gualtieri, 2008; Stauch and Lehmkuhl, 2010; Barr and
Clark, 2011, 2012). This knowledge gap motivated the mapping
survey of Margold and Jansson (2011) together with field-based
efforts, reported here, to establish quantitative age control on glaciations
in central Transbaikalia with 10
Be-derived exposure dating.
2.2. Field sites
Our study centres on the Kodar Mountains, the highest massif in
the region, which currently hosts over 30 small cirque glaciers
(Preobrazhenskiy, 1960; Shahgedanova et al., 2002, 2011; Stokes
et al., 2013; Osipov and Osipova, 2014). We collected samples for
exposure dating from boulders on moraine complexes in the Chara
Depression and in the River Vitim valley (Table 1; Fig. 4). In the
Chara Depression samples derive from: (i) inner (V-12-MM-01 to
03) and outer (V-12-MM-04 to 06) ridges of the Apsat River
moraine complex (Figs. 4 and 5a); (ii) inner (V-12-MM-12 to 14)
and outer (V-12-MM-10 and 11) ridges of the Leprindo Lakes
moraine complex (Figs. 4 and 5b); and (iii) the Ikabja moraine,
located on the southern boundary of the Chara Depression and
formed by a glacier flowing from the Udokan Range (V-12-MM-07
to 09; Figs. 4 and 5c). In the Vitim River valley, northwest of the
Kodar Mountains, we sampled a moraine formed by a glacier
emanating from the Lake Oron valley (GLV-11 and 13; Figs. 4 and
6a), and the Baznyi moraine formed by a glacier in the Amalyk
River valley (V-12-MM-16 to 17; Figs. 4 and 6b) just northeast of a
Fig. 3. Mapped glacial geomorphology redrawn from Margold and Jansson (2011). Glacially modified terrain (U-shaped valleys, cirques, ar^etes) is shaded purple, moraines in dark
red, deltas in orange, indistinct deltas in yellow, and glacial lakes in blue. White dashed lines indicate the branching valley glaciers on whose moraines we collected samples for 10
Be
exposure dating. A circle in the highlands northeast of Bodaybo as well as two ellipses on the northeast slopes of the Kodar Mountains indicate zones of areal glacial scouring (after
Margold and Jansson, 2011). Note two distinct generations of moraines in the main valley draining the Udokan Mountains to the south-southeast (marked by white stars). (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174164
giant bedrock spillway associated with a glacial lake outburst flood
(see Margold et al., 2011). All samples were of medium to coarsegrained
granite lithology with the exception of GLV-13, which
consisted of mica-schist.
With the aim of extending the knowledge of glaciations farther
afield, we sampled a glacially eroded bedrock surface (coarsegrained
granite) on a hillslope above the Vitim River opposite the
town of Mamakan (V-12-MM-15; Fig. 7aec) near Bodaybo. We also
collected a sample for OSL analysis from glaciolacustrine sediments
deposited behind a moraine in the Tamarak valley downstream of
the Aglan-Yan, a minor range south of Bodaybo (TK-12-01; Figs. 2
and 7a, d-f). At this site, a ~20 m high section of upwardscoarsening
diamicton is exposed in a road-side gravel pit (see
Fig. 7f). At the eastern end of the section, several metres of stratified
sediment grade up from clayey-silts at base to silts with intercalated
sandy beds. We attribute this sequence to a minor proglacial
lake, which formed when the drainage of a tributary valley coming
from the NNE was blocked by a glacier lobe in the Tamarak River
valley (see Fig. 7d).
3. Methods
3.1. Cosmogenic 10
Be analyses
Samples for exposure dating were collected with a hammer and
chisel, and sample coordinates and elevations were recorded via
hand-held GPS (Table 1). Samples GLV-11 and GLV-13, collected in
2011, were processed at Deutsches GeoForschungsZentrum GFZ in
Potsdam. Quartz was separated and cleaned following procedures
based on Kohl and Nishiizumi (1992) and 10
Be was extracted using
ion chromatography following procedures described by von
Blanckenburg et al. (2004) and von Blanckenburg et al. (1996),
using approximately 250 mg 9
Be carrier solution prepared from
phenakite mineral, with 9
Be concentration of 372.5 ± 3.5 ppm.
10
Be/9
Be ratios of these two samples were measured at the University
of Cologne's accelerator mass spectrometer laboratory
(Dewald et al., 2013). The 10
Be/9
Be ratio of the full procedural blank
prepared with these samples was 0.8 ± 0.3 Â 10À15
and blank
corrected 10
Be/9
Be ratios obtained for GLV-11 and GLV-13 were
2.5 ± 0.1 Â 10À13
and 2.4 ± 0.1 Â 10À13
, respectively. The 10
Be/9
Be
ratios obtained for the two samples were normalised to the 2007
KNSTD standard (see Nishiizumi et al., 2007 and Balco et al., 2008
for more details).
The V-12-MM series of samples, collected in 2012, were processed
at the Australian Nuclear Science and Technology Organisation
(ANSTO) using a combination of standard HF/HNO3 methods
after Child et al. (2000) and hot phosphoric acid (Mifsud et al.,
2013). All samples were spiked with a 10
Beefree solution of
approximately 300 mg 9
Be prepared from a beryl crystal solution at
1080 (±0.7%) ppm. Full procedural chemistry blanks were prepared
from the same beryl and gave a final mean 10
Be/9
Be ratio of
13.2 ± 1.5 Â 10À15
(n ¼ 4, mean of two procedural blanks). Blank
corrected 10
Be/9
Be ratios ranged between 0.8 and 8.0 Â 10À13
with
analytical errors ranging from 2 to 6%. The 10
Be/9
Be ratios were
measured at the ANSTO ANTARES accelerator (Fink and Smith,
2007) and were normalised to the NIST SRM-4325 standard with
a revised nominal 10
Be/9
Be ratio of 2.790 Â 10À11
. Errors for the final
10
Be concentrations (atoms/g) were calculated by summing in
quadrature the statistical error for the AMS measurement, 2% for
reproducibility, and 1% for uncertainty in the Be spike
Table 1
Cosmogenic 10
Be sample data and modelled surface exposure ages.
Location
Sample code
Latitude/Longitude
(
N/
E)
Elevation
(m above
sea level)
Sample
thicknessa
(cm)
Topographic
shielding
factor
Height of
boulder
above
ground (cm)
Quartzb
Mass
(g)
Be carrier
mass
(mg)
10
Be/9
Bec
(x10À15
)
10
Be Concentrationc
(x103
atoms g À1
SiO2)
Agec, d
(ka)
Apsat moraine inner crest
V-12-MM-01 56.98518/118.15501 936 3 1 80 33.20 0.345 88.4 ± 4.3 61.40 ± 3.29 6.3 ± 0.5
V-12-MM-02 56.98513/118.15450 926 3 1 100 30.60 0.344 203.5 ± 7.0 152.90 ± 6.27 16.0 ± 1.1
V-12-MM-03 56.98460/118.15321 927 3 1 120 29.47 0.340 281.8 ± 6.3 217.26 ± 6.87 22.8 ± 1.4
Apsat moraine outer crest
V-12-MM-04 56.99836/118.30656 805 2 1 60 35.11 0.338 177.8 ± 5.3 114.39 ± 4.27 13.2 ± 0.8
V-12-MM-05 57.00034/118.31067 782 2.5 1 110 32.13 0.341 94.5 ± 3.9 67.03 ± 3.12 7.9 ± 0.5
V-12-MM-06 57.00052/118.31419 767 1 1 60 29.41 0.341 79.7 ± 5.0 61.75 ± 4.11 7.3 ± 0.6
Leprindo moraine inner crest
V-12-MM-12 56.60032/117.4916 1108 2 1 140 24.84 0.346 166.2 ± 9.7 154.72 ± 9.70 13.8 ± 1.1
V-12-MM-13 56.59451/117.49396 1142 3 1 40 12.43 0.346 77.6 ± 5.0 144.33 ± 9.83 12.6 ± 1.1
V-12-MM-14 56.60017/117.49263 1108 3 1 100 32.40 0.343 156.9 ± 9.0 111.00 ± 6.85 10.0 ± 0.8
Leprindo moraine outer crest
V-12-MM-10 56.66717/117.67627 954 2 1 30 34.31 0.339 300.8 ± 8.3 198.66 ± 7.05 20.2 ± 1.3
V-12-MM-11 56.66788/117.67362 940 3 0.999 30 22.97 0.343 35.1 ± 3.4 35.03 ± 3.50 3.6 ± 0.4
Ikabja moraine
V-12-MM-07 56.93895/118.69913 970 2.5 1 50 32.52 0.340 278.5 ± 10.8 194.58 ± 8.70 19.5 ± 1.3
V-12-MM-08 56.93906/118.69975 970 3 1 50 40.96 0.344 407.3 ± 11.8 228.60 ± 8.38 23.0 ± 1.5
V-12-MM-09 56.9398/118.70044 966 3.5 1 50 38.55 0.338 795.6 ± 14.3 466.26 ± 13.38 47.5 ± 2.8
Oron delta moraine
GLV-11 57.15510/116.3327 436 2 0.998 80 37.62 0.736 248.0 ± 9.9 120.62 ± 4.93 19.7 ± 1.3
GLV-13 57.15567/116.33258 427 3 0.998 60 36.03 0.734 238.0 ± 9.3 120.79 ± 4.83 20.1 ± 1.3
Baznyi moraine
V-12-MM-16 57.43035/116.57731 515 2 0.999 120 25.9 0.331 125.2 ± 3.0 107.03 ± 3.51 16.2 ± 1.0
V-12-MM-17 57.43903/116.5736 513 2 0.999 250 39.9 0.332 139.6 ± 3.9 77.53 ± 2.76 11.8 ± 0.7
Mamakan glacially eroded surface
V-12-MM-15 57.82583/114.03742 394 3 0.999 e 28.2 0.332 152.3 ± 4.4 119.63 ± 4.38 20.5 ± 1.3
a
The tops of all samples were exposed at the surface.
b
A density of 2.7 g cmÀ3
was used based on the granitic composition of the samples.
c
All uncertainties are reported at the 1-sigma level. Blank corrected 10/9 ratios. See text for details on level of blanks.
d
Exposure ages were calculated with the Cosmic-Ray Produced Nuclide Systematics (CRONUS) Earth online calculator (Balco et al., 2008) version 2.2; constants 2.2.1;
(http://hess.ess.washington.edu/), using 10
Be SLHL production rate of Heyman (2014). Full details of the cosmogenic 10
Be analyses and exposure age calculations are provided
in Section 3.1.
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174 165
concentration.
Exposure ages were calculated using the CRONUS online
calculator (Balco et al., 2008; version 2.2; constants 2.2.1) and are
reported here using the time-dependent CRONUS Lm4
production
rate scaling of Lal (1991) and Stone (2000), with a reference spallation
production rate of 3.94 ± 0.20 atoms gÀ1
yrÀ1
(Heyman,
2014). The 10
Be half-life of 1.387 ± 0.012 Myr (Chmeleff et al.,
2010; Korschinek et al., 2010) was used in all calculations.
3.2. Optically stimulated luminescence analyses
Sample TK-12-01 was collected using a stainless-steel tube
hammered into the sediment profile and processed at the Stockholm
University luminescence laboratory. Sediment from the outer
part of the sample was used for gamma spectrometric measurements;
cosmic ray dose was calculated using present-day depth
and assuming average water content of 40 ± 5% during burial,
considering the fine-grained nature of the sediment and its position.
Material from the inner part of the sample was sieved
(63e100 mm) and chemically pre-treated (HCl, H2O2, Na-oxalate).
The quartz fraction was isolated using heavy liquids, and then
etched with 40% HF (60 min), followed by 10% HCl treatment
(>120 min) and re-sieving. Equivalent dose (De) was determined
(Risø DA-20) using a modified version of the SAR-protocol of
Murray and Wintle (2000), with preheating at 230 C for 10 s prior
to all OSL measurements (blue diode stimulation for 60 s at 125 C,
Hoya U340 detection filter).
4. Results
The full set of 10
Be analytical data and exposure ages are presented
in Table 1 grouped according to location. Exposure ages from
all sampled moraines display a relatively large spread (Table 1;
Fig. 8), often up to a factor of two, despite total external errors of
about ±7%. For the Apsat River moraine complex (Figs. 4 and 5a),
exposure ages from the inner moraine (V-12-MM-01, -02, -03)
range from 23 to 6 ka, whereas for the outer moraine (V-12-MM-
04, -05, -06) the range is 13 to 7 ka. A similar spread of 20 to 4 ka
was obtained for the outer Leprindo moraine (V-12-MM-10, -11;
Figs. 4 and 5b), whereas the ages for the inner Leprindo moraine (V-
12-MM-12, -13, -14) were a more consistent 14 to 10 ka. Boulders
sampled on the large Ikabja moraine (Figs. 4 and 5c) returned
exposure ages of 48 to 20 ka (V-12-MM-07, -08, -09). The two
boulders sampled on the small moraine directly in the Vitim River
valley west of Lake Oron (GLV-11, -13; Figs. 4 and 6a) yielded
overlapping ages (at 1s) of 20 ka. Two boulders sampled on the
Baznyi moraine in the Amalyk River valley (Figs. 4 and 6b) returned
ages of 16 and 12 ka (V-12-MM-16, -17, respectively). The single
bedrock sample collected from a glacially eroded surface across the
Vitim River from Mamakan (Fig. 7aec) yielded an exposure age of
21 ka.
For OSL sample TK-12-01, representing glaciolacustrine
Fig. 4. Highest mountain ranges of central Transbaikalia, the Kodar and Udokan, surrounding the Chara Depression, and the Vitim River skirting the northwest slopes: DEM-derived
image (Shuttle Radar Topography Mission, SRTM, data), showing dated moraines (red dots). Locations of Figs. 5 and 6 are marked by black rectangles. White arrows in the Syulban
valley (Kodar Mountains; upstream from Leprindo Lakes) indicate identified recessional moraines. See Fig. 2 for the location within the overall study area. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of this article.)
4
Adapting the Lal (1991) scaling scheme for paleomagnetic corrections. See
Balco et al. (2008) for more details.
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174166
Fig. 5. Chara Depression moraine complexes with sampling sites (red dots): Landsat image (LE71290212000275SGS00; channels 5, 3, 1) draped over SRTM data. (a) Apsat moraines,
(b) Leprindo Lakes moraines, (c) Ikabja moraine. Photographs to the right of the map panels show the sampled boulders. See Fig. 4 for the location of Fig. 5 map panels. Outlines of
the moraine ridges are marked by arrows. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
deposition behind a moraine in the Tamarak valley (Fig. 7a, d-f), 48
replicate measurements were carried out (2 mm aliquot size), and
32 of those passed the usual rejection criteria (Wintle and Murray,
2006). The distribution of De values was found to be relatively
complex, implying the presence of partial bleaching in some of the
aliquots. Using the Central Age Model results in an age of
20.6 ± 2.4 ka, whereas the Minimum Age Model (MAM, db ¼ 0.20)
yields 10.6 ± 1.7 ka (Table 2; cf. Galbraith et al., 1999). The true age
of deposition is likely to fall between these two scenarios, but it
would require a larger number of samples to better constrain the
age estimates.
5. Discussion
5.1. Interpreting 10
Be concentrations in glacial landforms
The 10
Be concentration measured in a boulder on a moraine
crest reliably records the timing of glacial retreat assuming that the
sampled surface: 1) does not contain cosmogenic nuclides inherited
from an earlier ice-free interval; 2) has been continuously
exposed sub-aerially since emergence from under ice and has not
been shielded by sediment, snow or vegetation cover; 3) has not
been repositioned or reoriented appreciably; and 4) has
experienced negligible erosion. The above factors fall into two
categories: processes that lead to overestimation of the exposure
age (1), and processes that make the modelled ages too young (2, 3,
and 4). Such biases have been discussed in various studies that
include distinctions as to which processes dominate a given study
area depending on the mode of glaciation, and geomorphic and
climatic setting (e.g., Gosse and Phillips, 2001; Fink et al., 2006;
Balco, 2011; Heyman et al., 2011). It has been shown that bedrock
surfaces are generally more prone to nuclide inheritance (Briner
and Swanson, 1998; Stroeven et al., 2002), whereas incomplete
exposure is more common for moraine boulders due to postdepositional
disturbances (Heyman et al., 2011). For either case,
surface erosion leads to loss of accumulated cosmogenic nuclides
and hence a younger apparent exposure age, although weathering/
erosion rate corrections (typically a few mm/ka) are generally minor
for postglacially exposed surfaces (e.g. Andre, 2002). A correction
for the cumulative effect of snow cover is difficult to gauge
reliably over 104
y timescales (Schildgen et al., 2005), as is the influence
of a forest canopy (Plug et al., 2007), though both will result
in underestimation of the true exposure age (Gosse and Phillips,
2001). Here we apply no corrections to our measured 10
Be concentrations
to account for erosion or shielding and thus all
modelled ages should be considered as minima.
Fig. 6. Northwest slopes of the Kodar Mountains with sampling sites (red dots): Landsat image (LE71290212000275SGS00; channels 5, 3, 1) draped over SRTM data. (a) Oron delta
moraine, (b) Baznyi moraine. The moraine ridges are marked by black arrows. Photographs to the right of the map panels show the sampled boulders. See Fig. 4 for the location of
Fig. 6 map panels. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174168
5.1.1. Glaciation in the high mountain ranges: Kodar and Udokan
The mapped geomorphology (Margold and Jansson, 2011), part
of which includes the moraines sampled for exposure dating, indicates
extensive mountain ice fields in the Kodar and Udokan
mountains (Figs. 3 and 4). The two distinct generations of moraines
on the southern slopes of the Udokan Mountains are an optimal
target for investigation (see Fig. 3), but regrettably these were out of
reach for logistical reasons. Instead, we focused on deciphering the
chronology of glacial limits in the Chara Depression and on the
northwest side of the Kodar Mountains. Whereas the glacial evidence
northwest of the Kodar Mountains appears to be more
complicated (we discuss this further in Section 5.1.2), the moraines
in the Chara Depression represent terminal glacier positions. These
massive moraines were formed by outlet glaciers of the Kodar and
Udokan ice fields and some feature two distinct moraine ridges
(Fig. 5 a, b). In addition, a series of recessional moraines occur
sporadically upstream of the terminal moraine complexes, for
example, in the Syulban River valley (Fig. 4).
As noted above, the 10
Be concentrations for multiple samples
from a given single moraine spread over a range that cannot be
explained by the far smaller analytical error for all the samples
within the group. For example, the three samples from the Ikabja
moraine (V-12-MM-7, -8, -9) give more than a twofold range in 10
Be
concentration yet each has a total analytical error <5% (Table 1).
Therefore, the observed age spread must result from geological
factors and some of the post-depositional processes mentioned
above are strongly implicated. Unfortunately, the small number of
samples measured for individual moraines does not allow for a
more detailed quantitative interpretation of the ages, such as that
suggested by Applegate et al. (2012). However, we can propose
Fig. 7. (a) Bodaybo area, with sampling sites V-12-MM-15 and TK-12-01 (red dots), DEM-derived image (SRTM data). (b) Oblique Google Earth view on the confluence of the Vitim
and Mamakan rivers and on the glacially modified slopes of the Vitim River valley. The site of sample V-12-MM-15 is drawn by a red dot and the viewing direction is indicated in
panel a. (c) Photograph of the sampled glacially-scoured surface (ice flow direction into the image), the sampling site is indicated by a white arrow behind the person for scale. (d)
Oblique Google Earth view of the Tamarak River valley with a site of OSL sample TK-12-01. The interpreted palaeo-configuration with a glacier lobe damming a minor lake is drawn
by a dashed line (ice margin) and white tint (glacial lake). (e) Detail on the section of laminated sediments at the time of OSL sampling. (f) Panoramic view on the Tamarak site
showing a road-side gravel pit with a ~20 m high section of upwards-coarsening diamicton. At the eastern end of the section, several metres of stratified sediment grade up from
clayey-silts at base to silts with intercalated sandy beds, which we interpret as a minor proglacial lake sequence (see panel d for palaeo-configuration) from which OSL sample TK-
12-01 was collected. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174 169
qualitative interpretations that allow an age model to be selected,
based on our judgement of the main active geomorphic processes.
If the large spread in 10
Be concentrations resulted from widespread
and varied nuclide inheritance (despite this being rather uncommon;
Heyman et al., 2011), then the youngest age for each deposit
would be the more reliable as it would contain the lowest proportion
of inheritance and therefore a minimum age model would
be most appropriate. However, in that case, half of the sampled
moraines would be of mid-Holocene age, which is highly implausible
considering the available information on past climate and
glacial chronologies from the wider region (Mackay, 2007; Shichi
et al., 2009; Tarasov et al., 2009; Arzhannikov et al., 2012).
Alternatively, we infer that processes causing age underestimation
have affected some of our samples. Several factors can be
invoked. First, the study area lies within zones of continuous and
discontinuous permafrost (Maurer, 2007); hence, permafrost dynamics
have quite possibly contributed to post-depositional
disturbance throughout the Late Glacial and Holocene. Moreover,
the moraines might have been ice-cored (and possibly still are) and
the buried ice may have experienced melting during the warmerthan-present
climate of the early Holocene (Tarasov et al., 2009)
resulting in alteration of the moraine morphology. Second, the
existence of a temporary glacial lake has been documented in the
Chara Depression (Enikeev, 2008, 2009; Margold and Jansson,
2011). The outer Apsat moraine crest has the lowest elevation of
all moraines sampled in the Chara Depression and we observed
glaciolacustrine sediments outcropping at the distal slope of the
moraine (57.00155 N, 118.32108 E). If a glacial lake submerged
this moraine and others within the Chara Depression, wave erosion
along shorelines could potentially modify and destabilize the
moraine crests, leaving them prone to degradation even after
drainage of the lake. Third, the degree of weathering among
sampled boulders and blocks was highly variable. Whereas some
surfaces appeared highly weathered, with weathering rinds and
certainly some loss of material, others were fresh, highly resistant
to chiselling and without noticeable alteration of the rock surface.
Nonetheless, no apparent relationship is discernible between the
degree of weathering and the modelled exposure ages. Although
variable erosion rates would result in scattered, depressed ages, any
reasonable range of erosion rates acting continuously over the past
20e40 kyrs would not on its own explain the excessive spread in
10
Be concentrations. Similarly, no relation is found between the
height above ground surface of the sampled boulders and their
modelled age, with the possible exception of sample V-12-MM-11,
which was among the lowest free-standing heights and volumes
and where the local surface morphology possibly indicates progressive
surface degradation (see Fig. 5b).
We apply a maximum age model to our moraine boulder
exposure ages to infer the timing of ice retreat. Considering the
oldest age measured on each moraine: at Leprindo Lakes the outer
moraine dates to ~20 ka (i.e., matching the LGM), and the inner
moraine crest dates to ~14 ka (Fig. 5b); at the Apsat River moraine
complex, all ages on the outer moraine crest must be rejected since
they would make it younger than the inner crest and the inner
moraine ridge is dated to ~23 ka (i.e., again consistent with LGM
age; Fig. 5a). Following the same logic, the Ikabja moraine (Fig. 5c)
would date to ~48 ka. However, here we argue that since sample V-
12-MM-09 is more than two-times older than all others in our
dataset, and the other two Ikabja samples fall within the LGM (~23
and ~20 ka), this anomalously old sample is interpreted to reflect
nuclide inheritance. Akin to the other two moraine complexes, the
Ikabja moraine is thereby also attributed LGM age. Moving from the
Chara Depression to the northwest slopes of the Kodar Mountains,
the moraine in the Vitim River valley west of Lake Oron (Fig. 6a)
matches with the LGM (~20 ka), while the Baznyi moraine (Fig. 6b)
is of a Late Glacial age (~16 ka).
Our data illustrate the difficulty with conducting exposure age
dating in active permafrost areas (cf. Zech et al., 2005; Stroeven
et al., 2014). In opting for a maximum age model, we suggest our
data support a glacial advance between 23 and 20 ka, commensurate
with global LGM timing (Yokoyama et al., 2000; Clark et al.,
2009), plus a possible Late Glacial readvance after ~16 ka. Our
mapping indicates that glaciers in the Kodar Mountains reached
lengths of over 100 km (Fig. 3) at the LGM. This contrasts with the
limited glacier extents reconstructed for the LGM in the regions
farther east (Stauch et al., 2007; Stauch and Gualtieri, 2008; Stauch
and Lehmkuhl, 2010) and also with the south-eastern sector of the
Barents-Kara Ice Sheet, which did not reach continental Siberia
during OIS-2 (Svendsen et al., 2004).
5.1.2. Glaciation in the Bodaybo area
Whereas our data from the Kodar and Udokan mountains provide
strong evidence for an LGM ice advance in the high massifs of
Transbaikalia, the few data available from the Bodaybo area of the
Vitim valley to the north-east (Fig. 7) allow only for some speculations
on the broader regional glaciations. Margold and Jansson
(2011) interpreted the Vitim valley as glacially modified for a substantial
distance downstream of its passage through the high
mountain region of the Northern-Muysk, Delyun-Uran, and Kodar
Fig. 8. Modelled 10
Be exposure ages ±1 s external uncertainty for the five moraine
complexes. Note that sample V-12-MM-09 (42.7 ± 3.8 ka) is not shown.
Table 2
Summary OSL dating information (sample TK-12-01), with estimated water content (W) and measured concentration of dose-rate relevant elements (K, Th, U) used for
calculation of the dose rate (D). Number of replicate measurements (n) and observed overdispersion (od.) as well as equivalent dose (De) for both CAM and MAM are listed
together with the resulting ages.
Latitude Longitude Depth
(cm)
Elevation
(m)
Grain-size
(mm)
W
(%)
K
(%)
Th
(ppm)
U
(ppm)
D
(Gy kaÀ1
)
n od. De
(Gy)
Age
(ka)
57.4578 114.0901 300 630 63e100 40 ± 5 2.22 ± 0.11 3.53 ± 0.16 1.29 ± 0.13 2.05 ± 0.15 32 0.49 42.2 ± 3.7 20.6 ± 2.4
MAM 21.6 ± 3.0 10.6 ± 1.7
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174170
mountains (Figs. 2 and 3). Our field observations confirmed the
glacial modification of the river valley and extended it downstream
of Bodaybo where we observed bedrock glacial striations preserved
beneath a thin sediment cover (57.846 N, 114.154 E; Fig. 9) near
the junction of the Bodaybo and Vitim rivers.
The glacially-scoured hillslopes above the Vitim River near
Mamakan show signs of local weathering and exfoliation, though
our sampling site was carefully chosen from an intact glaciallyscoured
bedrock surface (Fig. 7c). It is likely that some surface
weathering has occurred since glacial striations were not observed,
and we cannot exclude the possibility that the sampled surface was
for some time shielded by colluvium, which would lower its
average 10
Be production rate. In any case, our modelled exposure
age of 21 ka is difficult to reconcile with the radiocarbon chronology
indicating that the Vitim River valley at Bodaybo was ice-free at
the time (see Section 2.1; Ineshin and Teten'kin, 2010). For ice to fill
the Vitim valley downstream of Bodaybo a substantial glaciation is
required, most likely in the form of an ice cap extending from the
high mountains over the highlands east and south of Bodaybo
(Fig.10a). The existence of this ice cap at some prior stage is evident
from the regional pattern of glacial modification, with areal
scouring observed in the highlands north of the Vitim River (see
Fig. 3; Margold and Jansson, 2011), and has been previously discussed
in the literature (Obruchev, 1923, 1932; Logatchev, 1974;
Zolotarev, 1974; Enikeev, 2009; Margold and Jansson, 2011).
Our OSL date from the Tamarak River valley (TK-12-01; Table 2,
Fig. 7a, d-f) indicates that a glacier ~20 km in length descended
from the Aglan-Yan Range at ~21 ka. Peaking at 1836 m asl, the
Aglan-Yan Range is substantially lower than the higher massifs to
the south, though it still bears evidence of mountain glaciation.
Based on this single OSL age, we speculate that the Aglan-Yan Range
and similar minor ranges to the north of the high mountain crux of
Transbaikalia were heavily glaciated at the LGM (Fig. 10).
5.2. Comparison with previous glacial chronologies for
Transbaikalia
Notwithstanding the debate concerning the multiplicity of
earlier glaciations in Transbaikalia (see Section 2.1), our dating
captures the OIS-2 glaciation only. According to the available
regional climate records, derived from pollen and Lake Baikal diatoms,
the coldest part of OIS-2 spans 26e20 ka (Müller et al., 2014)
and thus coincides with the global LGM (Yokoyama et al., 2000;
Clark et al., 2009). This period was not only colder but also drier
than the preceding OIS-3 (Prokopenko et al., 2001; Mackay, 2007).
Cold conditions remained until at least 14.5 ka (Shichi et al., 2009),
followed by warmer and possibly wetter than present-day conditions
at 14e13 ka (Shichi et al., 2009; Tarasov et al., 2009). Cooling
then returned between 12.8 and 11.6 ka, during the Younger Dryas
Chronozone (Mackay, 2007; Shichi et al., 2009; Tarasov et al., 2009).
Our data indicate that extensive glaciation existed during the
coldest part of OIS-2 in the high mountains of Transbaikalia with a
system of branching valley glaciers feeding from the Kodar
Mountains into the Vitim valley and into the Chara Depression,
which also received glaciers from the Udokan Range (Figs. 3 and 9).
Whereas solid evidence exists for large LGM mountain ice fields on
the highest massifs, the paucity of chronological constraints
beyond the high mountains invites a more speculative view. The
geomorphology is in clear support of an ice-cap glaciation to the
northwest of the Kodar Mountains that filled the Vitim valley with
ice for several hundreds of km, but the timing of this event is not
firmly established. We note that the suggestions of a Middle or
early Late Pleistocene glaciation from Logatchev (1974) and
Zolotarev (1974) match the more recent glacial chronologies from
the wider region that advocate diminishing ice extents leading up
to and including the LGM (Svendsen et al., 2004; Gillespie et al.,
2008; Stauch and Gualtieri, 2008; Arzhannikov et al., 2012).
A crude estimate of LGM equilibrium line altitude (ELA) using a
toe-to-headwall altitude ratio of 0.5 (Benn and Lehmkuhl, 2000)
coupled with the elevation of the lowest preserved lateral moraines
indicates a NW-SE altitudinal trend in ELA consistent with the
precipitation gradient. ELA grew in height from ~975 m asl in the
Aglan-Yan Range, to ~1500 m asl on the southern slopes of the
Udokan Range (Fig. 10b). Indeed, for the region northwest of the
Kodar Mountains an ELA at ~1000 m asl quite possibly provided
sufficient accumulation area for an ice cap to develop. Nonetheless,
given that we have just one date in favour of an LGM ice cap capable
of supporting a Vitim valley glacier beyond Bodaybo, and that this
date conflicts with the existing 14
C dates from the area (albeit via
potentially unreliable 14
C measurements on bone; e.g., Stafford
et al., 1987), we defer to future work on the question of the
regional ice-cap style glaciation.
5.3. A Late Glacial readvance?
A final question that remains difficult to resolve with our data
concerns the Late Glacial readvances and their extent. Notwithstanding
Mackay's (2007, p. 198) view that “the transition from the
end of the last glacial into the Holocene (Termination 1) was rapid
in Lake Baikal (only c. 4 ka)”, climate reconstructions elsewhere
indicate that conditions remained cold until ~14.5 ka, then cooled
again after warming briefly at 14.5e13 ka, coincident with the
Fig. 9. Glacial striations on a sediment covered bedrock surface (57.846 N, 114.154 E)
near the junction of the Bodaybo and Vitim rivers. See Fig. 7a for the location.
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174 171
Bølling-Allerød interstadial (Shichi et al., 2009; Tarasov et al.,
2009). Unfortunately, the large spread in our moraine boulder
exposure ages makes it difficult to provide a reliable conclusion on
the presence of a Late Glacial ice re-advance in the study area. With
this caveat stated, following a maximum age model for our moraines
(excepting the inheritance in V-12-MM-09) yields an LGM
age for the inner Apsat, outer Leprindo, Ikabja, and Oron moraines,
whereas the Baznyi and inner Leprindo moraines are attributed to a
Late Glacial advance. Interestingly, the latter two ‘Late Glacial’
moraines attest to an ice extent similar to that of the ‘LGM’ moraines.
We note that a similar pattern has recently been reported
from the Khangai Mountains of Mongolia (Fig. 1), where an ice readvance
at 17e16 ka reached within 1e2 km of the local LGM limits
(Rother et al., 2014). In addition, evidence of minor still-stands or
re-advances of glaciers higher in the mountains exists in the form
of recessional moraines in some mountain valleys (Fig. 4). Such
sites remained out of reach during our 2011 and 2012 field expeditions,
leaving much further work yet to be done.
6. Conclusions
We applied cosmogenic 10
Be exposure dating to glacial landforms
with the aim of investigating the glacial history of central
Transbaikalia. A set of eighteen 10
Be exposure ages were determined
from five moraine complexes emanating from the Kodar and
Udokan mountains. Our results indicate a major glaciation
involving extensive mountain ice fields in the highest massifs
during OIS-2, coincident with the global LGM. The relatively large
Fig. 10. (a) Schematic reconstruction of LGM ice extent (white) in central Transbaikalia; note that except for the southeast flanks of the Kodar Mountains the ice extent shown is
speculative. Large valley glaciers were fed from a hypothesised ice field that spanned the Kodar and lower mountains to the north with a similar ice field mapped also in the Udokan
Mountains. Evidence of a more extensive, ice-cap type glaciation (white tint) occurs northwest of the highest massifs, but the timing of this glaciation remains unconfirmed (see
main text). Dashed black line indicates vertical transect AeA0 pictured in panel b. (b) Reconstructed approximate elevation of the equilibrium line altitude (ELA) across the NW-SE
transect through the study area (marked by black arrows). Note the successively higher ELAs corresponding to a precipitation gradient between the wetter NW portion and the
drier, more continental SE portion of the study area.
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174172
spread among the exposure ages is most likely due to incomplete
exposure caused by post-depositional disturbance of the moraines.
Whereas two of the investigated moraines date to the Late Glacial
and minor recessional moraines do occur higher in the mountains,
the issue of incomplete boulder exposure compromises our ability
to specify the precise timing of a Late Glacial re-advance in the
region.
Strong evidence exists for an even more extensive ice-cap type
glaciation northwest of the Kodar Mountains. We argue, on the
basis of OSL-dated glacigenic sediments, that even the lower
mountains to the north of the highest massifs were heavily glaciated
at the LGM. Although a single 10
Be exposure age in support of
the ice cap also indicates an LGM age, this conflicts with existing
radiocarbon ages recording ice-free conditions at the Vitim River
near Bodaybo at the time (Ineshin and Teten'kin, 2010). A suggested
pre-OIS-2 timing for this extensive regional glaciation would be
consistent with the trend of successively smaller glaciations since
OIS-6, previously reconstructed in northern and northeastern
Eurasia (Svendsen et al., 2004; Stauch and Gualtieri, 2008;
Arzhannikov et al., 2012). Suffice to say that compelling evidence
exists for large ice fields in the mountains of Transbaikalia at the
LGM, but the timing of an even more extensive, regional ice-cap
glaciation has yet to be established.
Acknowledgements
We thank Anton Brichevskiy for his help during the 2012 field
season and our thanks extend to the staff of the Vitim Nature
Reserve. We further thank Steven A. Binnie for arranging the
measurement of samples GLV-11 and GLV-13 at the University of
Cologne AMS laboratory. Finally, we thank Julie Brigham-Grette and
an anonymous reviewer for their constructive comments on an
earlier version of the manuscript.
This work was supported by funding from the Swedish Society
for Anthropology and Geography, the Bolin Centre for Climate
Research, the Royal Swedish Academy of Sciences (FOA12Althin-
034), Stockholm University, and Stiftelse YMER-80 to Martin Margold;
and by funding from the Australian Institute of Nuclear Science
and Engineering (ALNGRA12021P) and the Australian
Research Council (DP130104023) to John Jansen.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.quascirev.2015.11.018.
References
Andre, M.-F., 2002. Rates of postglacial rock weathering on glacially scoured outcrops
(Abisko-Riksgr€ansen area, 68
N). Geogr. Ann. 84 A, 3e4.
Applegate, P.J., Urban, N.M., Keller, K., Lowell, T.V., Laabs, B.J.C., Kelly, M.A.,
Alley, R.B., 2012. Improved moraine age interpretations through explicit
matching of geomorphic process models to cosmogenic nuclide measurements
from single landforms. Quat. Res. 77, 293e304.
Arzhannikov, S.G., Braucher, R., Jolivet, M., Arzhannikova, A.V., Vassallo, R.,
Chauvet, A., Bourles, D., Chauvet, F., 2012. History of late Pleistocene glaciations
in the central Sayan-Tuva Upland (Southern Siberia). Quat. Sci. Rev. 49, 16e32.
Astakhov, V.I., 2013. Pleistocene glaciations of Northern Russiaea modern view.
Boreas 42, 1e24.
Balco, G., 2011. Contributions and unrealized potential contributions of cosmogenicnuclide
exposure dating to glacier chronology, 1990e2010. Quat. Sci. Rev. 30,
3e27.
Balco, G., Stone, J.O., Lifton, N.A., Dunai, T.J., 2008. A complete and easily accessible
means of calculating surface exposure ages or erosion rates from 10
Be and 26
Al
measurements. Quat. Geochronol. 3, 174e195.
Barr, I.D., Clark, C.D., 2011. Glaciers and climate in Pacific far NE Russia during the
last glacial maximum. J. Quat. Sci. 26, 227e237.
Barr, I.D., Clark, C.D., 2012. An updated moraine map of Far NE Russia. J. Maps 8,
431e436.
Bazarov, D.B., 1986. The Cenozoic of Cisbaikalia and Western Transbaikalia. Nauka,
Novosibirsk (in Russian).
Bazarov, D.B., Rezanov, I.N., Budaev, R.C., Imethenov, A.B., et al., 1981. The Geomorphology
of Northern Cisbaikalia and the Stanovoe Range. Nauka, Moscow
(in Russian).
Benn, D.I., Lehmkuhl, F., 2000. Mass balance and equilibrium-line altitudes of glaciers
in high-mountain environments. Quat. Int. 65e66, 15e29.
Bigg, G.R., Clark, C.D., Hughes, A.L.C., 2008. A last glacial ice sheet on the Pacific
Russian coast and catastrophic change arising from coupled iceevolcanic
interaction. Earth Planet. Sci. Lett. 265, 559e570.
Brigham-Grette, J., 2001. New perspectives on Beringian quaternary paleogeography,
stratigraphy, and glacial history. Quat. Sci. Rev. 20, 15e24.
Briner, J.P., Swanson, T.W., 1998. Using inherited cosmogenic 36
Cl to constrain glacial
erosion rates of the Cordilleran ice sheet. Geology 26, 3e6.
Child, D., Elliott, G., Mifsud, C., Smith, A.M., Fink, D., 2000. Sample processing for
earth science studies at ANTARES. Nucl. Instrum. Methods Phys. Res. Sect. B
Beam Interact. Mater. Atoms 172, 856e860.
Chmeleff, J., von Blanckenburg, F., Kossert, K., Jakob, D., 2010. Determination of the
10
Be half-life by multicollector ICP-MS and liquid scintillation counting. Nucl.
Instrum. Methods Phys. Res. Sect. B 268, 192e199.
Clark, P.U., Dyke, A.S., Shakun, J.D., Carlson, A.E., Clark, J., Wohlfarth, B.,
Mitrovica, J.X., Hostetler, S.W., McCabe, A.M., 2009. The last glacial maximum.
Science 325, 710e714.
Dewald, A., Heinze, S., Jolie, J., et al., 2013. CologneAMS, a dedicated center for
accelerator mass spectrometry in Germany. Nucl. Instrum. Methods Phys. Res. B
294, 18e23.
Enikeev, F.I., 2008. The Late Cenozoic of northern Transbaikalia and paleoclimates of
southern East Siberia. Russ. Geol. Geophys. 49, 602e610.
Enikeev, F.I., 2009. Pleistocene glaciations in the East Transbaikalia and the
Southeast of Middle Siberia. Geomorfologiya 40, 33e49 (In Russian).
Fink, D., McKelvey, B., Hambrey, M.J., Fabel, D., Brown, R., 2006. Pleistocene
deglaciation chronology of the Amery Oasis and Radok Lake, Northern Prince
Charles mountains, Antarctica. Earth Planet. Sci. Lett. 243, 229e243.
Fink, D., Smith, A., 2007. An inter-comparison of 10
Be and 26
Al AMS reference
standards and the 10
Be half-life. Nucl. Instrum. Methods Phys. Res. Sect. B Beam
Interact. Mater. Atoms 259, 600e609.
Galbraith, R.F., Roberts, R.G., Laslett, G.M., Yoshida, H., Olley, J.M., 1999. Optical
dating of single and multiple grains of quartz from Jinmium rock shelter,
northern Australia: Part I, experimental design and statistical models.
Archaeometry 41, 339e364.
Geniatulin, R.F., 2000. The Encyclopedia of Transbaikalia: Chita Region. Nauka,
Novosibirsk (In Russian).
Gillespie, A.R., Burke, R.M., Komatsu, G., Bayasgalan, A., 2008. Late pleistocene
glaciers in Darhad Basin, Northern Mongolia. Quat. Res. 69, 169e187.
Gosse, J.C., Phillips, F.M., 2001. Terrestrial in situ cosmogenic nuclides: theory and
application. Quat. Sci. Rev. 20, 1475e1560.
Heyman, J., 2014. Paleoglaciation of the Tibetan Plateau and surrounding mountains
based on exposure ages and ELA depression estimates. Quat. Sci. Rev. 91, 30e41.
Heyman, J., Stroeven, A.P., Harbor, J.M., Caffee, M.W., 2011. Too young or too old:
evaluating cosmogenic exposure dating based on an analysis of compiled
boulder exposure ages. Earth Planet. Sci. Lett. 302, 71e80.
Ineshin, E.M., Teten’kin, A.V., 2010. Human and Environment in the North of Baikalian
Siberia in Late Pleistocene. Archaeological site Bol’shoi Iakor’ I, Nauka,
Novosibirsk (In Russian).
Kohl, C., Nishiizumi, K., 1992. Chemical isolation of quartz for measurement of in
situ-produced cosmogenic nuclides. Geochimica CosmochimicaActa 6,
3583e3587.
Kolomiets, V.L., 2008. Paleogeography and quaternary terrace sediments and
complexes, intermontane basins of Prebaikalia (Southeastern Siberia, Russia).
Quat. Int. 179, 58e63.
Komatsu, G., Arzhannikov, S.G., Arzhannikova, A.V., Ershov, K., 2007a. Geomorphology
of subglacial volcanoes in the Azas Plateau, the Tuva Republic,
Russia. Geomorphology 88, 312e328.
Komatsu, G., Arzhannikov, S.G., Arzhannikova, A.V., Ori, G.G., 2007b. Origin of glacialefluvial
landforms in the Azas Plateau volcanic field, the Tuva Republic,
Russia: role of iceemagma interaction. Geomorphology 88, 352e366.
Korschinek, G., Bergmaierb, A., Faestermanna, T., Gerstmannc, U.C., Kniea, K.,
Rugela, G., Wallnerd, A., Dillmanna, I., Dollingerb, G., Lierse von Gostomskie, Ch,
Kossertf, K., Maitia, M., Poutivtseva, M., Remmer, A., 2010. A new value for the
half-life of 10
Be by heavy ion elastic recoil detection and liquid scintillation
counting. Nucl. Instrum. Methods Phys. Res. Sect. B 268, 187e191.
Kropotkin, P.A., 1876. Studies on the ice age. Pap. Russ. Geogr. Soc. 7, 1e837 (In
Russian).
Krivonogov, S.K., Takahara, H., 2003. In: Kamata, N. (Ed.), Late Pleistocence and
Holocence Environmental Changes Recorded in the Terrestrial Sediments and
Landforms of Eastern Siberia and North Mongolia, Proceedings of International
Symposium of the Kanazawa University 21st-Century COE Program, vol. 1,
pp. 30e36.
Kulchitskiy, A.A., Skovitina, T.M., Ufimtsev, G.F., 1995. The ability to quickly flood the
bottom of the Muya basin by a collapsing ravine in the Parama Canyon of the
Vitim. Environ. Aspects Theor. Appl. Geomorphol. Proc. Int. Conf. “III Shchukin
Read.” 136e137 (In Russian).
Kulchitskiy, A.A., Panychev, V.A., Orlova, L.A., 1990. Late Pleistocene deposits of the
Muya-Kuanda depression and their rate of accumulation. In: 7th Union Conference
on Quaternary Research “Quaternary Period: Research Methods, Stratigraphy
and Ecology”, Tallin, pp. 112e113 (In Russian).
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174 173
Lal, D., 1991. Cosmic ray labeling of erosion surfaces: in situ nuclide production rates
and erosion models. Earth Planet. Sci. Lett. 104, 424e439.
Lehmkuhl, F., Klinge, M., Stauch, G., 2004. The extent of Late Pleistocene glaciations
in the Altai and Khangai mountains. In: Ehlers, J., Gibbard, P.L. (Eds.), Developments
in Quaternary Sciences. Elsevier, pp. 243e254.
Lehmkuhl, F., Klinge, M., Stauch, G., 2011. Chapter 69-The extent and timing of late
pleistocene glaciations in the Altai and neighbouring mountain systems. In:
Jürgen Ehlers, P.L.G., Philip, D.H. (Eds.), Developments in Quaternary Sciences.
Elsevier, pp. 967e979.
Lehmkuhl, F., Zander, A., Frechen, M., 2007. Luminescence chronology of fluvial and
aeolian deposits in the Russian Altai (Southern Siberia). Quat. Geochronol. 2,
195e201.
Logatchev, N.A., 1974. Sayan-baikal Upland. In: Florensov, N.A. (Ed.), The Highlands
of Cisbaikalia and Transbaikalia. Nauka, Moscow (In Russian).
Mackay, A.W., 2007. The paleoclimatology of Lake Baikal: a diatom synthesis and
prospectus. Earth-Science Rev. 82, 181e215.
Mangerud, J., Jakobsson, M., Alexanderson, H., Astakhov, V., Clarke, G.K.C.,
Henriksen, M., Hjort, C., Krinner, G., Lunkka, J.P., M€oller, P., Murray, A.,
Nikolskaya, O., Saarnisto, M., Svendsen, J.I., 2004. Ice-dammed lakes and
rerouting of the drainage of Northern Eurasia during the last Glaciation. Quat.
Sci. Rev. 23, 1313e1332.
Margold, M., Jansson, K.N., 2011. Glacial geomorphology and glacial lakes of central
Transbaikalia, Siberia, Russia. J. Maps 7, 18e30.
Margold, M., Jansson, K.N., Stroeven, A.P., Jansen, J.D., 2011. Glacial Lake Vitim, a
3000-km3
outburst flood from Siberia to the Arctic Ocean. Quat. Res. 76,
393e396.
Maurer, J., 2007. Atlas of the Cryosphere. National Snow and Ice Data Center. Digital
media, Boulder, Colorado USA.
Mifsud, C., Fujioka, T., Fink, D., 2013. Extraction and purification of quartz in rock
using hot phosphoric acid for in situ cosmogenic exposure dating. Nucl. Instrum.
Methods Phys. Res. Sect. B Beam Interact. Mater. Atoms 294 (C), 203e207.
Müller, S., Tarasov, P.E., Hoelzmann, P., Bezrukova, E.V., Kossler, A., Krivonogov, S.K.,
2014. Stable vegetation and environmental conditions during the last glacial
maximum: new results from Lake Kotokel (Lake Baikal region, Southern Siberia,
Russia). Quat. Int. 348, 14e24.
Murray, A.S., Wintle, A.G., 2000. Luminescence dating of quartz using an improved
single-aliquot regenerative-dose protocol. Radiat. Meas. 32, 57e73.
Nishiizumi, K., Imamura, M., Caffee, M.W., Southon, J.R., Finkel, R.C., McAninch, J.,
2007. Absolute calibration of 10
Be AMS standards. Nucl. Instrum. Methods Phys.
Res. B 259, 403e413.
Obruchev, V.A., 1923. Olekma-vitim Gold-bearing Region. In: The Geological Review
of Gold-bearing Regions of Siberia. Lenzoloto, Moscow (In Russian).
Obruchev, V.A., 1929. On the glaciation of the central Vitim mountain region. Geogr.
Bull. 6 (4e6), 42e45 (In Russian).
Obruchev, V.A., 1932. A Geological Sketch of Cisbaikalia and the Lena Area. Essays on
the Geology of Siberia. The Academy of Sciences of the USSR, Leningrad.
Osadchiy, S.S., 1979. Limno-glacial conditions and the issues of correlation of the
Pleistocene formations in the depressions of the Stanovoe Highlands, the history
of lakes in the USSR in the late Cenozoic. In: Materials for the 5th All-union
Symposium, Part 2 (Central Siberia, Cisbaikalia, Transbaikalia, Yakutia, the Far
East). Irkutsk, pp. 122e126 (In Russian).
Osadchiy, S.S., 1982. On the Problem of the Relation of Glacial and Fluvial Epochs on
the Territory of Northern Transbaikalia, Late Cenozoic History of Lakes in the
USSR. Nauka, Novosibirsk (In Russian).
Osipov, E. Yu, Osipova, O.P., 2014. Mountain glaciers of southeast Siberia: current
state and changes since the Little ice age. Ann. Glaciol. 55, 167e176.
Plug, L.J., Gosse, J.C., McIntosh, J.J., Bigley, R., 2007. Attenuation of cosmic ray flux in
temperate forest. J. Geophys. Res. 112, F02022.
Preobrazhenskiy, V.S., 1960. Kodarskiy lednikovyy rayon (Zabaykal’ya) ([The
Kodarsky glacier region (Transbaikal)]). Nauka. Akademiia Nauk SSSR, Moscow
(Russian with English summary).
Prokopenko, A.A., Karabanov, E.B., Williams, D.F., Kuzmin, M.I., Khursevich, G.K.,
Gvozdkov, A.A., 2001. The detailed record of climatic events during the past
75,000 yrs BP from the Lake Baikal drill core BDP-93-2. Quat. Int. 80e81, 59e68.
Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C.,
Buck, C.E., Cheng, H., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P.,
Haflidason, H., Hajdas, I., Hatte, C., Heaton, T.J., Hoffmann, D.L., Hogg, A.G.,
Hughen, K.A., Kaiser, K.F., Kromer, B., Manning, S.W., Niu, M., Reimer, R.W.,
Richards, D.A., Scott, E.M., Southon, J.R., Staff, R.A., Turney, C.S.M., van der
Plicht, J., 2013. IntCal13 and Marine13 radiocarbon age calibration curves
0e50,000 Years cal BP. Radiocarbon 55, 1869e1887.
Rother, H., Lehmkuhl, F., Fink, D., Nottebaum, V., 2014. Surface exposure dating
reveals MIS-3 glacial maximum in the Khangai mountains of Mongolia. Quat.
Res. 82, 297e308.
Schildgen, T.F., Phillips, W.M., Purves, R.S., 2005. Simulation of snow shielding
corrections for cosmogenic nuclide surface exposure studies. Geomorphology
64, 67e85.
Shahgedanova, M., Mikhailov, N., Larin, S., Bredikhin, A., 2002. The mountains of
Southern Siberia. In: Shahgedanova, M. (Ed.), The Physical Geography of
Northern Eurasia. Oxford University Press, Oxford, pp. 314e349.
Shahgedanova, M., Popovnin, V., Aleynikov, A., Stokes, C.R., 2011. Geodetic mass
balance of Azarova glacier, Kodar mountains, Eastern Siberia, and its links to
observed and projected climatic change. Ann. Glaciol. 52, 129e137.
Shichi, K., Takahara, H., Krivonogov, S.K., Bezrukova, E.V., Kashiwaya, K.,
Takehara, A., Nakamura, T., 2009. Late Pleistocene and Holocene vegetation and
climate records from Lake Kotokel, central Baikal region. Quat. Int. 205, 98e110.
Stafford, T., Jull, A.T., Brendel, K., Duhamel, R.C., Donahue, D., 1987. Study of bone
radiocarbon dating accuracy at the university of Arizona NSF accelerator facility
for radioisotope analysis. Radiocarbon 29, 24e44.
Stauch, G., Gualtieri, L., 2008. Late quaternary glaciations in northeastern Russia.
J. Quat. Sci. 23, 545e558.
Stauch, G., Lehmkuhl, F., 2010. Quaternary glaciations in the Verkhoyansk mountains,
Northeast Siberia. Quat. Res. 74, 145e155.
Stauch, G., Lehmkuhl, F., Frechen, M., 2007. Luminescence chronology from the
Verkhoyansk mountains (North-Eastern Siberia). Quat. Geochronol. 2,
255e259.
Stokes, C.R., Shahgedanova, M., Evans, I.S., Popovnin, V.V., 2013. Accelerated loss of
alpine glaciers in the Kodar Mountains, South-eastern Siberia. Glob. Planet.
Change 101, 82e96.
Stone, J.O., 2000. Air pressure and cosmogenic isotope production. J. Geophys. Res.
105, 23753e23759.
Stroeven, A.P., Fabel, D., Harbor, J., H€attestrand, C., Kleman, J., 2002. Quantifying the
erosional impact of the Fennoscandian ice sheet in the Tornetr€ask-Narvik
corridor, Northern Sweden, based on cosmogenic radionuclide data. Geogr.
Ann. 84A, 275e287.
Stroeven, A.P., Fabel, D., Margold, M., Clague, J.J., Xu, S., 2014. Investigating absolute
chronologies of glacial advances in the NW sector of the Cordilleran ice sheet
with terrestrial in situ cosmogenic nuclides. Quat. Sci. Rev. 92, 429e443.
Stuiver, M., Reimer, P.J., 1993. Extended 14C database and revised CALIB radiocarbon
calibration program. Radiocarbon 35, 215e230.
Svendsen, J.I., Alexanderson, H., Astakhov, V.I., Demidov, I., Dowdeswell, J.A.,
Funder, S., Gataullin, V., Henriksen, M., Hjort, C., Houmark-Nielsen, M.,
Hubberten, H.W., Ingolfsson, O., Jakobsson, M., Kjaer, K.H., Larsen, E.,
Lokrantz, H., Lunkka, J.P., Lysa, A., Mangerud, J., Matiouchkov, A., Murray, A.,
Moller, P., Niessen, F., Nikolskaya, O., Polyak, L., Saarnisto, M., Siegert, C.,
Siegert, M.J., Spielhagen, R.F., Stein, R., 2004. Late quaternary ice sheet history of
Northern Eurasia. Quat. Sci. Rev. 23, 1229e1271.
Tarasov, P.E., Bezrukova, E.V., Krivonogov, S.K., 2009. Late Glacial and Holocene
changes in vegetation cover and climate in southern Siberia derived from a 15
kyr long pollen record from Lake Kotokel. Clim. Past. 5, 285e295.
Ufimtsev, G.F., Skovitina, T.M., Kulchitskiy, A.A., 1998. Rockfall-dammed lakes in the
Baikal region: evidence from the past and prospects for the future. Nat. Hazards
18, 167e183.
von Blanckenburg, F., Belshaw, N., O'Nions, R., 1996. Separation of Be-9 and
cosmogenic Be-10 from environmental materials and SIMS isotope dilution
analysis. Chem. Geol. 129, 93e99.
von Blanckenburg, F., Hewawasam, T., Kubik, P.W., 2004. Cosmogenic nuclide evidence
for low weathering and denudation in the wet, tropical highlands of Sri
Lanka. J. Geophys. Res. Earth Surf. 109 (F3), F03008.
Wintle, A.G., Murray, A.S., 2006. A review of quartz optically stimulated luminescence
characteristics and their relevance in single-aliquot regeneration dating
protocols. Radiat. Meas. 41, 369e391.
Yokoyama, Y., Lambeck, K., De Deckker, P., Johnston, P., Fifield, L.K., 2000. Timing of
the last Glacial maximum from observed sea-level minima. Nature 406,
713e716.
Zech, R., Glaser, B., Sosin, P., Kubik, P.W., Zech, W., 2005. Evidence for long-lasting
landform surface instability on hummocky moraines in the Pamir mountains
(Tajikistan) from 10
Be surface exposure dating. Earth Planet. Sci. Lett. 237,
453e461.
Zhao, J., Yin, X., Harbor, J.M., Lai, Z., Liu, S., Li, Z., 2013. Quaternary glacial chronology
of the Kanas river valley, Altai mountains, China. Quat. Int. 311, 44e53.
Zolotarev, A.G., 1974. The Baikal-Patom Upland. In: Florensov, N.A. (Ed.), The
Highlands of Cisbaikalia and Transbaikalia. Nauka, Moscow (In Russian).
M. Margold et al. / Quaternary Science Reviews 132 (2016) 161e174174