Quaternary Science Reviews 30 (2011) 1173e1184
Contents lists available at ScienceDirect
Quaternary Science Reviews
journal homepage: www.elsevier.com/locate/quascirev
Holocene vegetation and climate histories in the eastern Tibetan Plateau: controls
by insolation-driven temperature or monsoon-derived precipitation changes?
Yan Zhao a, *,1, Zicheng Yu b,1, Wenwei Zhao a
a
b
MOE Key Laboratory of Western China’s Environmental System, Research School of Arid Environment and Climate Change, Lanzhou University, Lanzhou 730000, China
Department of Earth and Environmental Sciences, Lehigh University, Bethlehem, PA 18015, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 June 2010
Received in revised form
15 February 2011
Accepted 23 February 2011
Available online 21 March 2011
The climates on the eastern Tibetan Plateau are strongly influenced by direct insolation heating as well as
monsoon-derived precipitation change. However, the moisture and temperature influences on regional
vegetation and climate have not been well documented in paleoclimate studies. Here we present a welldated and high-resolution loss-on-ignition, peat property and fossil pollen record over the last 10,000
years from a sedge-dominated fen peatland in the central Zoige Basin on the eastern Tibetan Plateau and
discuss its ecological and climatic interpretations. Lithology results indicate that organic matter content
is high at 60e80% between 10 and 3 ka (1 ka ¼ 1000 cal yr BP) and shows large-magnitude fluctuations in
the last 3000 years. Ash-free bulk density, as a proxy of peat decomposition and peatland surface
moisture conditions, oscillates around a mean value of 0.1 g/cm3, with low values at 6.5e4.7 ka, reflecting
a wet interval, and an increasing trend from 4.7 to 2 ka, suggesting a drying trend. The time-averaged
mean carbon accumulation rates are 30.6 gC/m2/yr for the last 10,000 years, higher than that from many
northern peatlands. Tree pollen (mainly from Picea), mostly reflecting temperature change in this alpine
meadow-forest ecotonal region, has variable values (from 3 to 34%) during the early Holocene, reaches
the peak value during the mid-Holocene at 6.5 ka, and then decreases until 2 ka. The combined peat
property and pollen data indicate that a warm and wet climate prevailed in the mid-Holocene (6.5
e4.7 ka), representing a monsoon maximum or “optimum climate” for the region. The timing is
consistent with recent paleo-monsoon records from southern China and with the idea that the interplays
of summer insolation and other extratropical large-scale boundary conditions, including sea-surface
temperature and sea-level change, control regional climate. The cooling and drying trend since the midHolocene likely reflects the decrease in insolation heating and weakening of summer monsoons.
Regional synthesis of five pollen records along a southenorth transect indicates that this climate pattern
can be recognized all across the eastern Tibetan Plateau. The peatland and vegetation changes in the late
Holocene suggest complex and dramatic responses of these lowland and upland ecosystems to changes
in temperature and moisture conditions and human activities.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Holocene
Forest decline
Peatlands
Pollen
Peat
Summer monsoon
Insolation
Zoige Basin
Eastern Tibetan Plateau
1. Introduction
Summer monsoon intensity in East Asia has varied at multiple
timescales. During the Holocene it is hypothesized that the strongest
monsoon during the early Holocene was induced by peak summer
insolation (Kutzbach, 1981; Ruddiman, 2008). This hypothesis has
been confirmed in a general term by increasing empirical evidence
from cave deposits (Wang et al., 2005, 2008), lake sediment records
(e.g., Shen et al., 2005) and peat-based records (e.g., Hong et al.,
2003; Zhou et al., 2004). However, the timing of Holocene
* Corresponding author. Tel.: þ86 931 8912337; fax: þ86 931 8912330.
E-mail address: [email protected] (Y. Zhao).
1
These authors contributed equally.
0277-3791/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.quascirev.2011.02.006
monsoon maximum appears to vary in different regions and from
various records (e.g., Overpeck et al., 1996; Xiao et al., 2004; Jiang
et al., 2006; Griffiths et al., 2009; Yang and Scuderi, 2010; Yang
et al., 2010). For example, pollen records from Daihai Lake in the
monsoonal region of north China suggest a mid-Holocene monsoon
maximum (Xiao et al., 2004). The speleothem records from IndoPacific region show that the maximum monsoon occurred after 7 ka
(1 ka ¼ 1000 cal yr BP) after sea level rose and stabilized (Griffiths
et al., 2009). Further analysis based on Chinese cave records
suggests that the lowest oxygen isotope values that were recorded
during the Holocene may not reflect maximum precipitation, as the
maximum precipitation intensity seemed to have occurred much
later after the summer insolation maximum in the region between
Dongge and Heshang caves (Hu et al., 2008). The roles of sea level in
1174
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
tropical southeastern Asia (Griffiths et al., 2009) and extratropical
boundary conditions (ice sheet and sea-surface temperature) have
been proposed to explain the inconsistencies in timing of monsoon
maximum (Overpeck et al., 1996; Herzschuh, 2006). Furthermore, it
is unclear whether temperature or precipitation changes have been
dominant features of the observed regional climate and environmental changes.
The eastern Tibetan Plateau is strongly influenced by Asian
summer monsoons, including both the East Asian and Indian
monsoons. However, its high-altitude (>3000 m above sea level)
settings should make it also sensitive to direct insolation heating and
other extratropical influences, like many high-latitude regions, in
addition to monsoon-driven precipitation changes. Peatlands
(alpine marshes) on the eastern Tibetan Plateau are the largest
highland peatland in the world. Multiple proxy data from peat-core
records would potentially provide information to evaluate the
relative importance of monsoon-derived precipitation and
temperature changes during the Holocene. Palynological data from
peatlands can be used to reconstruct regional vegetation and its
spatial and temporal patterns. Peat properties, including organic
matter content and degree of peat decomposition, can be used to
infer hydrological conditions on the peatland surface. Combining
these proxies would allow us to evaluate local and regional vegetation changes and their potential climate controls. During the last
decade, several studies have been published on peatlands from this
region, mostly for pollen analysis and regional vegetation and
climate reconstructions (e.g., Yan et al., 1999; Joosten et al., 2008;
Zhou et al., 2010). However, few studies have used peat properties
to infer paleoclimatic changes (but see Zhou et al., 2002). Furthermore, the regional patterns of Holocene vegetation changes and
their climate controls are still poorly understood.
There is also a debate about the relative importance of climate
and human activities in causing the observed vegetation changes,
especially forest decline, on the Tibetan Plateau during the midand late Holocene (e.g., Miehe et al., 2009; Schlütz and Lehmkuhl,
2009; Herzschuh et al., 2010). For example, studies from southcentral Tibet by Miehe et al. (2009) and Schlütz and Lehmkuhl
(2009) suggest that human activities, especially grazing, have had
significant impacts on vegetation since as early as 8000 years ago.
However, on the basis of pollen-based precipitation reconstructions
at a site in the northeastern Tibetan Plateau, Herzschuh et al. (2010)
concluded that monsoon-induced precipitation changes could
explain the forest decline during the last 6000 years and that
human activities are not necessary. Obviously additional records
from other regions in the Tibetan Plateau would provide useful
information on this topic.
In this study, we present a Holocene record of peat properties and
fossil pollen data from a rich fen peatland in the central Zoige Basin.
The objectives of this study were (1) to reconstruct regional vegetation and peat accumulation histories using multiple proxy data
from a peat core; (2) to evaluate the relative importance and influence of the monsoon-driven precipitation and insolation-driven
temperature on peatland dynamics, regional vegetation and climate
changes; and (3) to investigate the potential different responses and
sensitivities of regional vegetation along a southenorth transect in
the eastern Tibetan Plateau to Holocene temperature and moisture
changes.
2. Study region and site
The Zoige Basin is a low-relief plateau in the eastern Tibetan
Plateau at 32100 e34100 N latitude and 101450 e103 250 E longitude, with an altitude of ca 3350e3450 m above sea level (Figs. 1A
and 2A). The basin contains a long lake sedimentary deposit, going
back to 800 ka (Chen et al., 1999). The peatland area is ca 4500 km2,
Fig. 1. Location map and climate settings. A. Satellite image map of the Tibetan Plateau
and surrounding regions showing the locations of paleo study sites discussed in the
text: 1. Zoige peatland (core ZB08-C1, this study; red large dot); 2. Hongyuan peatland
sites (Yan et al., 1999; Zhou et al., 2010); 3. Zoige core RM (Shen and Tang, 1996); 4.
Dalianhai Lake (Cheng, 2006); 5. Qinghai Lake (Shen et al., 2005); 6. Dongge Cave
(Wang et al., 2005); and 7. Heshang Cave (Hu et al., 2008). B and C. The mean annual
temperature and mean annual precipitation of the Tibetan Plateau, respectively (from
Institute of Geography, 1990). The rectangle in B and C shows the study region as
detailed in Fig. 2.
with average peat depth of 2e3 m and a maximum peat depth of up
to 9e10 m (Thelaus, 1992; Joosten et al., 2008).
Mean annual precipitation (MAP) at nearby Zoige meteorological station (at 3439 m a.s.l.) is 648.5 mm for the period 1971e2000.
Most precipitation falls as rain during the summer months
(JuneeSeptember; Fig. 3C), owing to the influence of the Asian
summer monsoons. Mean annual temperature (MAT) is 1.1 C, with
July temperature of 10.8 C and January temperature of 10.2 C.
We also plotted the climate diagrams from meteorological stations
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
1175
Fig. 2. Study sites and vegetation. A. DEM of the eastern Zoige Basin at 32e34.5 N and 102e104 E. B. Simplified vegetation types of the same region as in A (modified from Hou,
2001): forest dominated by Picea with some Abies and Betula; shrubland by Salix and Rhododendron, alpine meadow by Kobresia spp., and peatland (alpine marshes) by Carex
muliensis. Also shown are core locations: core ZB08-C1 in the central Zoige Basin, Hongyuan peatland (HY) in the southern Zoige Basin, and core RM in the northern Zoige Basin.
C. Satellite image of the study region showing the peatlands (in dark color) and coring site (image source: Google Earth).
along the SEeNW transect near other paleo study sites (Fig. 3). MAP
shows a decreasing trend from 769 mm in Hongyuan to 379 mm in
Gangcha, near Qinghai Lake, while MAT shows a similar decrease
from >1 C in the Zoige Basin to 0.3 C near Qinghai Lake, despite
its much lower altitude (1971 m a.s.l.).
The basin is primarily covered by alpine meadows dominated by
Kobresia spp., other plants in the sedge family (Cyperaceae), Artemisia, Poaceae, and Ranunculaceae, with abundant peatlands
(“alpine marshes”) in low-lying broad valleys between hills and
rivers (see Fig. 2B and C). The peatlands are dominated by sedges
including Carex muliensis and Kobresia humilis (see Fig. 4). Subalpine meadows occur on the slopes of mountains near the basin at
3400e3800 m a.s.l., mainly composed of sedge and grass species,
e.g., Kobresia setchwanensis, Clinelymus nutans, Poa partensis,
together with some species in Asteraceae, Ranunculaceae and
Fabaceae (Shen, 2003). The plateau is today dominated by pasture
land inhabited by nomads. Some grazing weeds (e.g., Boraginaceae,
Bistorta, Caragana, Potentilla and Stellera) can be found inside the
pastures. The surrounding mountains, especially to the east and
south, are covered by scattered forests at up to 4000 m a.s.l, mainly
composed of Picea asperata, Picea wilsonii, Picea purpurea, Abies
faxoniana, Pinus densata, Betula platyphylla and Quercus liaotunggensis (Shen, 2003; Zhou et al., 2010), and by shrublands
dominated by Salix and Rhododendron (Hou, 2001). Temperature is
dominant climate controls of major vegetation types in this ecotonal region between forest and alpine meadow (Table 1).
The study site is located just north of a highway (Route #209)
between the towns of Zoige and Tanggor, about 20 km east of the
first major bend of the Yellow River (Fig. 2C). The coring site is at the
southwestern end of one of many SWeNE trending valley peatlands in this region (Fig. 2C). The local peatland vegetation is
dominated by sedges (C. muliensis and K. humilis). There are no
1176
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
croplands near the peatland, but there are limited cattle grazing
around relatively dry edge of the peatland at the present. The
nearest forest stands are about 30 km to the east of the study site
(Fig. 2B).
3. Methods
3.1. Field core collection
We collected a 650-cm long peat core (core ZB08-C1) at the
southern end of a large inter-valley peatland in the Zoige Basin
(coring site coordinates at 33 270 N and 102 380 E, with an elevation
of 3467 m a.s.l.; Fig. 2C) in June 2008 using a Macaulay peat corer.
The top 28 cm was cut as a monolith using a bread knife. The water
table was near the peatland surface at the time of coring (Fig. 4B).
The core section at 28e100 cm was not recovered due to high water
content and loose peat materials (Fig. 5). Each 50-cm long core
segment was wrapped in plastic wraps and stored and transported
in split PVC pipes.
3.2. Laboratory subsampling and analysis
The peat core was cut into contiguous 1-cm-thick slices. Nine
samples of microscopic charcoal particles (>125 mm in size) and
sedge seeds were picked for accelerator mass spectrometry (AMS)
radiocarbon dating at Keck AMS Lab at University of CaliforniaIrvine (Table 2). All dates were calibrated to calendar years before
present (0 BP ¼ 1950 AD) with the program CALIB Rev. 5.0.1 using
IntCal04 calibration data set (Reimer et al., 2004). The ageedepth
model was established based on the 3rd polynomial curve
(Fig. 6A).
Volumetric subsamples of w2 cm3 were used for loss-on-ignition (LOI) analysis. Sequential combustion at 500 C and 1000 C
was used to estimate organic matter and carbonate contents,
respectively (Dean, 1974). Dry weight and sample volume were
used to calculate bulk density at every 1 cm depth. Ash-free
(organic matter) bulk density was calculated from the measurements of bulk density and organic matter content. Apparent carbon
accumulation rates were calculated using calibrated AMS ages, ashfree bulk density measurements and carbon content of peat organic
matter in peatlands (using 52% C in peat organic matter as in Vitt
et al., 2000).
A total of 153 pollen subsamples of w0.5 cm3 in volume were
taken at 4-cm intervals. The subsamples were processed for pollen
following standard procedure (Fægri and Iversen, 1989), including
HCl, KOH, HF and acetolysis treatments, and fine sieving to remove
clay-sized particles. Pollen sums were usually >400 terrestrial
pollen grains. Known amount of Lycopodium spores were added at
the beginning of pollen preparation to help estimate pollen
concentration.
3.3. Data analysis
Fig. 3. Climate diagrams from four meteorological stations in the eastern Tibetan
Plateau showing monthly temperature and precipitation. A. Gangcha, Qinghai
(37 200 N, 100 080 E; elevation of 1971 m above sea level); B. Maqu, Gansu (34 000 N,
102 050 E; 3471 m elevation); C. Zoige, Sichuan (33 350 N, 102 580 E; 3439 m elevation);
D. Hongyuan, Sichuan (32 480 N, 102 330 E; 3492 m elevation). All data were from
climate normals for the period 1971e2000. MAP: mean annual precipitation; MAT:
mean annual temperature.
Pollen zonation was based on CONISS (Grimm, 1987) using the
dominant pollen taxa from core ZB08-C1. Total tree pollen
percentages at core ZB08-C1 and each of four other pollen
sequences from the eastern Tibetan Plateau (Hongyuan peatland in
the southern Zoige Basin, core RM from the northern Zoige Basin,
Dalianhai Lake, and Qinghai Lake; see Table 3 for site information)
were resampled into 500-year bins by averaging tree pollen
percentages of all samples in a binned interval. We used the same
pollen sum, including Cyperaceae pollen, for tree pollen percentage
calculations, as Cyperaceae pollen mostly comes from alpine
meadow (see below). We recalculated tree pollen percentages from
published pollen diagram at Hongyuan peatland (Zhou et al., 2010)
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
1177
Fig. 4. Ground photos of the studied peatland near the coring site of core ZB08-C1. A. Broad view of the study peatland and surrounding hilly landscape in the Zoige Basin, looking
northward. B. Close-up view of the coring site, which is dominated by sedges (mostly Carex muliensis).
by including Cyperaceae pollen in pollen sum. The binned tree
pollen percentages were then standardized to zero mean and unit
standard deviation. All standardized tree pollen curves were averaged to generate the composite regional tree pollen record, with
standard error as an estimate of errors and variations.
4. Results
4.1. Radiocarbon dates and core chronology
There is a dating reversal between two dates at 470 cm
(6420 50 14C BP) and 507 cm (4280 60 14C BP). We rejected the
date at 507 cm due to its small amount of carbon and apparently
too young age (Table 2). Eight accepted dates (in calendar ages)
used for ageedepth model are in order and fit the 3rd polynomial
curve within the dating and calibration errors (Fig. 6A). Chronology
indicates that the peat core covers the last 10,300 years. The
temporal sampling resolution is w15 years for each contiguous 1cm interval for lithology analysis and w65 years for fossil pollen
record. Ages based on this age model were used in the subsequent
discussion.
4.2. LOI analysis and peat lithology
The study site initiated as a shallow-water pond in the early
Holocene from 10.3 to 9.7 ka (650e580 cm), with carbonate
content of up to >40% and w50% organic matter (Fig. 6B, C). A
banded peat occurred at 9.7e3.4 ka (580e270 cm) with organic
matter of around 70% and <5% carbonate (Fig. 6). This banded peat
section has alternated light (well-preserved) and dark (highly
humified) peat layers (Fig. 5). The banding becomes weak after
4.5 ka (320 cm). A dark, highly humidified, massive (not banded)
peat layer occurred at 3.4e2.4 ka (270e230 cm). A well-preserved,
fresh-looking peat was present at 2.4e1.3 ka (230e170 cm). A
mineral sediment layer at 1.3e0.5 ka (170e110 cm) has very low
organic matter of <20% and >80% silicate. The top 110 cm peat (the
last 500 years) has the highest organic matter content of >80%.
Ash-free bulk density varies mostly around 0.1 g/cm3, with
millennial-scale oscillations. It shows the lowest values at
6.5e4.5 ka, except during the mineral-rich layer of the last 1500
years (Fig. 6E). The time-averaged mean of apparent peat carbon
accumulation rates is 30.6 gC/m2/yr, with slightly high rates around
7 ka and during the last 1000 years (Fig. 6F).
Table 1
Altitudinal vegetation distribution in the eastern Tibetan Plateau (from Shen and Tang, 1996).
Vegetation
Dominant taxa
Alpine desert steppe
Poaceae, Artemisia, Chenopodiaceae,
Asteraceae
Cyperaceae, Poaceae, Asteraceae
Cyperaceae, Poaceae
Picea, Abies
Alpine meadow
Sub-alpine meadow
Sub-alpine coniferous forest
Altitude distribution (m)
>4200
3600e4200
3300e3600
2100e3600
Annual temperature ( C)
Precipitation
(mm/yr)
<4
<500
4 to 0.7
0.7 to 1.5
1e5
500e600
600e750
600e1000
1178
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
Fig. 5. Core photos of core ZB08-C1 from the central Zoige Basin on the eastern Tibetan Plateau.
4.3. Fossil pollen results
We identified 26 pollen types in 153 samples from core ZB08-C1.
A summary percentage pollen diagram is shown in Fig. 7. The entire
pollen assemblages were dominated by Cyperaceae, with abundance ranging from w50% to >80%. Other major pollen types
include Picea, Pinus, Betula, Poaceae, Ranunaculaceae and Artemisia.
The percentage pollen diagram was divided into 3 pollen assemblage zones, based on stratigraphically constrained cluster analysis
(CONISS) and visual inspection (Fig. 7).
Zone ZB-1 (10.3e7.3 ka; 650e446 cm): Pollen assemblages were
dominated by Cyperaceae (mean of 70.2%, ranging from 50% to
>90%), with some Picea, Betula and Ranunaculaceae. Tree pollen
percentages fluctuated between 8.5% and 33.7%. A key feature of
this pollen zone is its highly fluctuating pollen abundance,
especially as illustrated by trees and Cyperaceae. This zone also
has the highest and fluctuating total pollen concentration
(Fig. 7).
Zone ZB-2 (7.3e3.6 ka; 446e288 cm): Pollen assemblages have
the lowest Cyperaceae pollen (mean of 64.5%), slightly high
Ranunaculaceae, and high and stable tree pollen (up to 35%).
Zone ZB-3 (3.6e0 ka; 288e0 cm): This zone was marked by
a substantial reduction in tree pollen, mainly Picea (<5%) and
Betula (<2%), and the highest and stable Cyperaceae pollen (up
to 96%). Two subzones were divided at 1.3 ka (173 cm), mainly
based on a slight increase in Picea, Poaceae and Ranunaculaceae
in subzone 3b.
5. Discussion
5.1. Local peatland and regional vegetation changes in the Zoige
Basin during the Holocene
The peatland at our study site experienced long-term and
millennial-scale changes since its initiation in the early Holocene.
The peatland initiated at 9.7 ka from terrestrialization (lake-infilling) process from a shallow pond, as indicated by high mineral and
calcareous sediments at the base of the core (Fig. 6C). This pond-topeatland transition is part of autogenic succession, which could have
been triggered by climate change (Vitt, 2006; Yu et al., 2009). The
peatland shows long-term stability until 6.5 ka, as indicated by high
and slightly increasing organic matter content (w70%) and average
ash-free bulk density (w0.1 g/cm3) (Fig. 6). Ash-free bulk density
Table 2
AMS radiocarbon dates from the Zoige peatland (core ZB08-C1) on the eastern Tibetan Plateau.
Lab number
Depth (cm)
Material dated
d13C (&)
14
UCIAMS-58979
UCIAMS-58980
UCIAMS-58981
UCIAMS-58982
UCIAMS-58983
UCIAMS-58984
UCIAMS-58985
UCIAMS-54697
UCIAMS-58986
140
210
270
360
430
470
507
577
640
Charcoal,
Charcoal
Charcoal,
Charcoal
Charcoal,
Charcoal
Charcoal,
Charcoal
Charcoal,
25.2
ea
27.2
27.3
e
e
e
e
e
945
1870
3160
4790
5970
6420
4280b
9030
8905
a
b
3 sedge seeds
1 charred wood
2 sedge seeds
4 sedge seeds
1 sedge seed
C date (yr BP)
Sample was too small for d13C analysis. For these samples, d13C value of 25& was assumed for correcting
Date was not used in the age model.
14
Error (yr)
Calibrated age (cal yr BP-2s range)
15
90
20
20
30
50
60
80
50
796e874
1591e1996
3355e3415
5475e5546
6729e6892
7266e7426
4788e4979
9908e10,304
9887e10,199
C dates.
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
1179
Fig. 6. Age model and lithology results from core ZB08-C1 in the central Zoige Basin. A. Age model of core ZB08-C1; B. Organic matter content; C. Carbonate content; D. Silicate
content; E. Ash-free bulk density (1-cm raw data as thin gray line, and 5-point smoothed data in bold black line); F. Carbon accumulation rates between paired age determinations.
reflects the degree of preservation of peat, as highly decomposed
peat is more compacted, denser and has higher bulk density (Yu
et al., 2003), and also likely reflects climate conditions, as a warm
and wet climate would increase productivity and quick peat burial,
producing low bulk density. Millennial-scale fluctuations in bulk
density may correlate with tree pollen abundance, and its highfrequency variations may correspond to the fine-scale color changes
(Fig. 5). The decreasing trend in organic matter after 6.5 ka (down to
33% at 2.4 ka) corresponds to long-term increase in bulk density and
long-term decrease in tree pollen, suggesting an increase in
decomposition and deforestation, likely induced by a drying and
cooling climate trend. At the beginning of this long-term trend,
a long-interval low bulk density at 6.5e4.7 ka started at the highest
tree pollen around 6.5 ka. A return to high organic matter content at
2.4e1.3 ka correlates with an increase and then decrease in bulk
density and the lowest tree pollen, perhaps reflecting a cool and wet
climate. The high mineral interval at 1.3e0.5 ka indicates a severe
disturbance of the peatland, as the clastic materials were likely
transported to the peatland by water or wind.
Three pollen assemblage zones from core ZB08-C1 show clear
changes in regional and local vegetation over the last 10,000 years in
the Zoige Basin. Modern surface pollen analysis indicates that
Cyperaceae pollen comes from sedge plants (Carex and Kobresia)
growing on both alpine meadows and lowland peatlands (Yu et al.,
2001; Shen et al., 2006; Herzschuh and Birks, 2010). Using an
extensive data set of 227 surface pollen samples from the eastern
and central Tibetan Plateau, including the Zoige Basin, Shen et al.
(2006) show that pollen assemblages from “upland” alpine
meadow routinely contain about 60% Cyperaceae pollen. Our limited
surface pollen samples from alpine meadow and peatlands in the
Zoige Basin show similar pattern (F.R. Li and Y. Zhao, unpublished
data). Unfortunately, we are unable to identify Cyperaceae pollen to
genus and species levels based on pollen morphology. However, on
the basis of these modern pollen assemblage studies and the fact
that our study site has remained to be a peatland throughout its
history, we assume that contribution of wetland sedge species to
Cyperaceae pollen in our pollen diagram has been relatively
constant, so the relative variations in Cyperaceae pollen in core
ZB08-C1 mostly reflect change in background pollen rain from
regional alpine meadow. Furthermore, owing to the poor dispersal
ability of Picea pollen (e.g., Sugita, 1993) and its close association
with parent plants as documented in surface pollen studies in the
eastern Tibetan Plateau (Yu et al., 2001; Lu et al., 2008) and elsewhere (Liu et al., 1999), most Picea pollen grains were derived from
local source in the watershed rather than long-distance transport,
especially during the periods with high Picea pollen abundance. Lu
et al. (2008) indicate that in the Tibetan Plateau the samples with
>20% Picea and Abies suggest the presence of sub-alpine coniferous
forest. Therefore, we interpret that major increase in total tree pollen
abundance, especially Picea, reflects either the establishment of local
tree populations in the hills around our study peatland (see Figs. 4
and 2C) or the forest edge from the northeast migrating closer to
Table 3
Site information of the five pollen records used for data synthesis from the eastern Tibetan Plateau.
Site
Latitude ( N)
Longitude ( E)
Altitude (m asl)
MAT ( C)
MAP (mm)
No. of
dates
Pollen sampling
resolution (yr)
Reference
Hongyuan peatland
Zoige ZB08-C1
Zoige core RM
Dalianhai Lake
Qinghai Lake
32 470
33 270
33 570
36 150
36 320
102 310
102 380
102 210
100 240
99 360
3505
3467
3401
2850
3200
1
1.1
0.9
3.3
0.7
700
650
705
300
250
31
8
3
8
6
175
65
150
80
60
Zhou et al., 2010
This study
Shen and Tang, 1996
Cheng, 2006
Shen et al., 2005; Herzschuh
et al., 2010
MAT: Mean annual temperature; MAP: mean annual precipitation.
1180
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
Fig. 7. Summary percentage pollen diagram from core ZB08-C1 in the central Zoige Basin on the eastern Tibetan Plateau. Open curves are 5 exaggerations for minor taxa. Analyst:
W.W. Zhao.
our study site (Fig. 2B), under favorable climate conditions. We do
not think that there have been major changes in pollen source areas
for most part of our record after the site became a fen peatland at
9.7 ka. Our fossil pollen record (Fig. 7) indicates that forest or
woodland with scattered trees established in the watershed, alternating with alpine meadows, from 10.3 to 7.3 ka, relatively stable
tree populations occurred at 7.3e3.6 ka, and major deforestation
occurred since 3.6 ka, except a brief recovery around 1 ka.
The multiple proxy record from core ZB08-C1 in the central
Zoige Basin reflects change in both temperature and moisture
conditions during the Holocene. The large-magnitude oscillations
in pollen abundance at 10.3e6.5 ka as reflected in total tree pollen
(Fig. 8B) indicate variable and multi-centennial-scale change in
temperature (and also moisture) conditions. Both Picea and Betula
are dominant trees, and both show similar long-term trends and
high-frequency fluctuations. Most favorable climate conditions,
likely warm and wet, occurred at 6.5e4.7 ka, for the highest tree
pollen, and likely local occurrence of tree populations around the
site, indicate a warm climate and the lowest bulk density indicates
moisture conditions on the peatland. From 4.7 to 2.4 ka, the
increasing bulk density and decreasing tree pollen suggest a drying
and cooling climate trend. The drying and reduced tree cover may
induce more wind-blown dust and mineral material to the peatland, causing decreasing organic matter content and increasing
silicate (Fig. 6B, D). The decrease in silicates from 4.7 to 2.4 ka is not
very likely caused by human activities, such as grazing, as its
decrease is gradual and following a smooth long-term trend. Also,
Ranunaculaceae and Fabaceae pollen, potentially containing
grazing indicator pollen types (e.g., Anemone-type, Trollius, Caragana; Schlütz and Lehmkuhl, 2009), shows decrease rather than
increase (see below for more discussion on human activities). The
lowest tree pollen, high organic matter content and decreasing bulk
density at 2.4e1.3 ka suggest a cool and moist condition, which
limits tree establishment but promotes preservation of peat. The
mineral layer and small peak in tree abundance around 1 ka
suggest a warm and wet climate condition, and extremely wet
condition might have caused flooding and erosion of mineral
material from the surrounding hills. The different responses of
Fig. 8. Synthesis of tree pollen percentages from study sites on the eastern Tibetan Plateau. A. Hongyuan peatland in the southern Zoige Basin (Zhou et al., 2010); B. Core ZB08-C1 in
the central Zoige Basin (this study); C. Core RM in the northern Zoige Basin (Shen and Tang, 1996); D. Dalianhai Lake (Cheng, 2006); E. Qinghai Lake (Shen et al., 2005), and F.
Synthesized tree pollen pattern from five regional pollen records, showing as means and standard errors of the 5-site means at 500-year binned intervals in standard deviations
(S.D.) units. Small black squares in each panel mark the dating points for each pollen record. See Fig. 1A for site locations.
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
landscape stability and erosion to warm and wet climates in the
mid-Holocene and late Holocene may reflect the different sensitivity of these systems to various degrees of deforestation.
5.2. Sensitivity of peatland dynamics and upland vegetation to
Holocene climate change in the eastern Tibetan Plateau
Our multi-proxy record from Zoige peatland site shows different
responses of upland vegetation and peatland development to
climate changes during the Holocene. Peatland initiation at 9.7 ka
from a pond at our study site suggests either an autogenic
succession through lake-infilling process or a response to early
Holocene climate warming. Basal peat dates from other sites in the
northern Zoige Basin and in Hongyuan also show an early Holocene
peatland initiation (e.g., Thelaus, 1992; Yan et al., 1999; Zhou et al.,
2010). Also, a well-dated peat record from Hongyuan appears to
show the peak peat accumulation around 10 ka (Zhou et al., 2010).
Hemispheric peat-core data syntheses from northern peatlands in
boreal and subarctic regions indicate that both peatland initiation
and carbon accumulation peaked in the early Holocene around
10 ka (MacDonald et al., 2006; Yu et al., 2009), which was attributed to the greatest insolation and climate seasonality (Yu et al.,
2010). Warm summers would stimulate plant production, and
cold winters would reduce decomposition of organic matter (Jones
and Yu, 2010). If the maximum peatland initiation and carbon
accumulation in the early Holocene in the Zoige Basin can be
further confirmed by additional dates and analysis, then it would
have significant implications for understanding fundamental
controls of peatland formation and dynamics by insolation seasonality, not only at high latitudes but also in high-altitude regions
at low latitudes.
A warm and wet climate in the mid-Holocene around 6.5 ka as
indicated by the maximum tree pollen and lowest bulk density
might have contributed to the slightly high carbon accumulation at
that time (Fig. 6F). The subsequent increase in bulk density and
decrease in tree pollen until 2.4 ka were caused by a drying and
possibly cooling climate trend, when the summer monsoon
weakens and summer insolation decreases. Since 2.4 ka it appears
that the peatland responded to climate change or other forcings in
a more dramatic and complex manner, which is still poorly understood and requires additional regional records to elucidate possible
causes. The most dramatic switch from high organic matter at
2.4e1.3 ka to low organic matter at 1.3e0.5 ka might have represented a non-linear threshold response to change in temperature
and moisture conditions and subsequent change in upland vegetation cover and landscape stability. A cool and moist climate at
2.4e1.3 ka caused the lowest tree cover and well-preserved peat,
while subsequent dramatic decrease in organic matter at 1.3e0.5 ka
might have been caused by flooding and upland erosion in a warm
and extreme wet climate.
Alternatively, human activities, especially grazing, might have
contributed to the intensified disturbance after 1.3 ka (Thelaus,
1992). Although we do not have direct palynological evidence for
human activities, the slight increase in Ranunaculaceae, Poaceae and
Asteraceae pollen after 1.3 ka suggest the initiation or increase in
grazing activities, as those family-level pollen types contain several
grazing indicators (Anemone-type, Trollius, and Cichoriodeae) as
claimed by Schlütz and Lehmkuhl (2009). In any case, our peat-core
record from the central Zoige Basin indicates there are no significant
human impacts on the study site before 1.3 ka.
Regional vegetation appears to show different response to
climate changes in the Holocene on the eastern Tibetan Plateau.
Along the south-to-north regional transect, the maximum tree
pollen abundance in the mid-Holocene varied from >80% in Hongyuan peatland, to less than 30% at our study site in the central Zoige
1181
Basin and 80% in the northern Zoige Basin, to <60% at Qinghai Lake
(Fig. 8). The sensitivity and nature of upland vegetation response to
climate change depend in part on the climate (temperature vs.
moisture) gradient and on the nature and proximity of surrounding
vegetation. For example, in the Zoige Basin vegetation change
between alpine meadow and forests mostly reflects change in
temperature, while around Qinghai Lake the changes between forest
and temperate steppe are most sensitive to moisture variations. A
vegetation shift from temperate steppe to alpine steppe during the
Holocene at Lake Zegetang in central Tibet was interpreted as caused
by climate cooling (Herzschuh et al., 2006). Also, the small-magnitude of tree pollen decline from 80% to 40% in the northern Zoige
Basin (core RM) was likely due to the close proximity of forest region
just northeast of the basin, causing less sensitive vegetation
response (Fig. 2B). In any case, most of these sites along the transect
show abrupt forest declines at 4e3 ka as indicated by the composite
regional tree pollen curve (Fig. 8F), indicating sensitive and nonlinear responses to gradual large-scale climate changes caused by
insolation-driven weakening in summer monsoons and neoglacial
cooling in the last several thousand years.
5.3. Holocene climate changes in the eastern Tibetan Plateau and
their large-scale controls
Holocene climates in the eastern Tibetan Plateau have experienced changes in both temperature and moisture conditions. The
eastern Tibetan Plateau has been strongly influenced by Asian
summer monsoons during the Holocene, which show maximum
intensity in the earlier Holocene and weakening since the midHolocene (e.g., Wang et al., 2005; Shao et al., 2006). Most previous
paleoclimate studies from this region often invoke change in
regional precipitation in response to large-scale monsoon circulation (e.g., Herzschuh et al., 2010; Zhou et al., 2010). However, here we
argue that temperature change has also played an important role in
causing regional vegetation change and has been an important
feature of regional climate change, especially in the southern and
humid part of the eastern Tibetan Plateau, including the Zoige Basin.
We base our argument on (1) the modern surface pollen studies
from the eastern Tibetan Plateau, (2) regional pollen data synthesis
along a transect from south to north on the eastern Tibetan Plateau,
and (3) broad-scale controls, directly and indirectly, of summer
insolation on high-latitude regional climate. We discuss each of
these lines of evidence in the following paragraphs.
In their correlation analysis of modern surface pollen assemblages
and climate data, Shen et al. (2006) found that annual precipitation
and summer temperature are two dominant climate parameters
controlling pollen assemblages. Furthermore, they found that Picea
pollen abundance shows stronger correlation with temperature than
precipitation and that Cyperaceae pollen has the strongest negative
correlation with temperature than any other pollen types, stronger
than correlations with precipitation. In another more extensive
modern pollen analysis of 857 samples, especially on Picea and Abies
distribution on the Tibetan Plateau, Lu et al. (2008) found that the
pollen abundance of these conifer trees is highly correlated with the
distribution of their parent trees, which is largely controlled by
elevation and temperature. In southern part of the eastern Tibetan
Plateau, temperature shows greater gradient than precipitation
(Fig. 1B and C), and coniferous forests prefer higher temperature than
alpine meadows, the other dominant vegetation types in the Zoige
Basin, while the favorable precipitation conditions have large overlap
between forests and meadows (Table 1). Thus, high tree pollen and
low Cyperaceae abundance should reflect warm as well as wet
climates.
The observed Holocene vegetation changes at core ZB08-C1
appears to show similar pattern with other pollen records from the
1182
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
eastern Tibetan Plateau. The sites from south to north include Hongyuan peatland at the southern edge of the Zoige Basin (Yan et al.,
1999; Zhou et al., 2010), core RM in the northern Zoige Basin (Shen
and Tang, 1996), Dalianhai Lake (Cheng, 2006) and Qinghai Lake
(Shen et al., 2005; Herzschuh et al., 2010), both lakes in the northeastern Tibetan Plateau (see Figs. 1A and 2A for site locations). All
these pollen records show peak tree pollen abundance at or near 6 ka
in the mid-Holocene (Fig. 8). Some of them also show highly variable
tree pollen percentages in the earlier Holocene, including especially
Hongyuan peatland (Zhou et al., 2010) and Dalinghai Lake (Cheng,
2006). Tree pollen or forest decline occurred at all sites since 6 ka,
but Hongyuan, RM and Dalinghai records show abrupt tree decline at
4e3.5 ka, as at core ZB08-C1. The differences between these records
could reflect either different vegetation histories (for example,
between Hongyuan and other two sites) or dating uncertainties (for
example, only one date available before 3.5 ka at Zoige core RM). In
the northeastern Tibetan Plateau, for example at Qinghai Lake, the
forest decline is interpreted as representing decrease in precipitation
in response to a weakening summer monsoon (Herzschuh et al.,
2010), as the lake is surrounded by extensive temperate steppe with
an annual precipitation range from <250 to >550 mm, more sensitive
to precipitation change. However, in the Zoige Basin, including
Hongyuan and Zoige sites, the vegetation changes between forests
and alpine meadows (Fig. 2B) should be more sensitive to temperature change as also documented in the modern surface pollen studies
(see above).
A mid-Holocene moisture maximum or “climate optimum” as
documented at our study site and other sites in the eastern Tibetan
Plateau is in agreement with recent monsoon precipitation reconstructions. For example, Hu et al. (2008) reconstructed precipitation
from two speleothem isotopic records in southwest China by differencing co-eval d18O values for the Dongge and Heshang caves,
about 600 km apart along the same moisture trajectory (see site
locations in Fig. 1A) and by removing secondary controls on d18O
(e.g., moisture source, moisture transport, non-local rainfall, and
temperature). The resulting Dd18O signal is directly controlled by the
amount of precipitation falling between two sites. Based on calibration with instrumental precipitation data, the reconstructed
precipitation shows a maximum at 6 ka, 8% higher than at the
present (Fig. 9D; Hu et al., 2008). This reconstruction is different
from earlier interpretation directly based on the d18O values from
individual sites (e.g., Dykoski et al., 2005; Wang et al., 2005), which
suggests the maximum monsoon intensity in the early Holocene
when the d18O values were lowest. Low oxygen isotope values at
either Dongge or Heshang caves in the early Holocene was probably
due to the remote vapor source, longer distance of moisture transport, and subsequent more depletion in oxygen isotope values, when
global sea level was lower (Cheng et al., 2009; Griffiths et al., 2009).
Insolation affects regional climate directly by radiative heating
and indirectly by changing atmospheric circulation. Summer insolation plays an important role in controlling the land-sea heating
contrast and monsoon intensity (Ruddiman, 2008), and peak
summer insolation in the early Holocene should cause the strongest
monsoon during the Holocene based on global model simulations
(Kutzbach, 1981). The different timings of monsoon maximum or
“climate optimum” in the eastern Tibetan Plateau and perhaps other
monsoonal regions from the insolation maximum likely reflect the
complex direct and indirect responses of regional climate to largescale climate controls. For example, Overpeck et al. (1996) attribute
the time lag of several thousand years in maximum monsoon
intensity in the AfricaneAsian monsoon region after peak summer
insolation to the result of slow-changing glacial boundary conditions (i.e., sea-surface temperatures and glacial ice sheets), retarding
the ability for the Tibetan Plateau to warm up with the gradual
increase in summer insolation. Low sea-surface temperature (SST) in
tropics and extratropics (Fig. 9G) during the early Holocene induced
by the remaining ice sheet might have affected the sea-land
temperature contrast and caused less moist availability and weak
monsoon intensity. The monsoon maximum or “climate optimum”
around 6.5 ka corresponds with the maximum SST, especially in the
North Atlantic (Fig. 9G). The decreasing summer insolation over the
later half of the Holocene (Fig. 9F) has caused the so-called neoglacial
cooling in many high-latitude regions (e.g., MacDonald et al., 2000;
Kaufman et al., 2004). Despite its subtropical latitude, the high
altitude of the Tibetan Plateau may make it to respond to summer
insolation in a similar manner as high-latitude regions, which might
explain the cooling since 6.6 ka. In addition to insolation influence
on the summer monsoon, low sea level in the early Holocene
Fig. 9. Regional and global correlations. A. Regional tree pollen pattern (standard deviations) from the eastern Tibetan Plateau (n ¼ 5); B. Tree pollen percentages from core ZB08-C1
(5-point smoothed curve); C. Ash-free bulk density of core ZB08-C1 (5-point smoothed curve); D. Reconstructed precipitation based on the difference of oxygen isotopes between
Dongge and Hesheng caves (Hu et al., 2008); E. Oxygen isotope record from Dongge Cave in southwest China (Dykoski et al., 2005); F. Summer insolation at 30 N and 60 N latitudes
(Berger and Loutre, 1991); G. PCA-1 score of sea-surface temperature (SST) deviations from the North Atlantic (Kaplan and Wolfe, 2006) and SST from the tropics (Rimbu et al.,
2004).
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
(Camoin et al., 1997; Peltier and Fairbanks, 2006) might have also
limited the availability of source moisture fueling the monsoon (e.g.,
Griffiths et al., 2009). Also, low sea level during the early Holocene
might have induced longer transport pathway of moisture and
therefore low oxygen isotope values, rather than great precipitation
amount (Wang et al., 2005). These analyses suggest that Holocene
sea-surface temperature and sea-level histories might have indirectly contributed to the mid-Holocene timing for the maximum
monsoon in the eastern Tibetan Plateau and other monsoon regions.
6. Conclusions
A new peat-core record from the central Zoige Basin in the
eastern Tibetan Plateau shows complex interactions between local
peatland development, upland vegetation and regional climate.
After peatland initiation at 9.7 ka, tree pollen and peat properties
indicate a highly variable and fluctuating local and regional environment before reaching optimum conditions at 6.5e4.7 ka. Forest
or tree population decline since the mid-Holocene corresponds
with increases in clastic sediment input and in peat decomposition,
suggesting a drying and cooling trend. Major large-magnitude
oscillations in the last 3 ka may reflect non-linear threshold
responses of peatland ecosystems to landscape stability induced by
changes in climate, vegetation or human activities.
Along a south-to-north transect in the eastern Tibetan Plateau,
temperature appears to play a major role in causing vegetation changes
between alpine meadows and forests in the south during the Holocene,
while precipitation might have been a dominant factor in semi-arid
region further north, causing vegetation shifts between forests and
temperate steppes. We emphasize the value and importance of
multiple proxy data in separating temperature and moisture changes
in interpreting vegetation and climate changes in the Holocene.
The monsoon maximum or “climate optimum” occurred at
6.5e4.7 ka at our study site and generally in the mid-Holocene from
other sites in the eastern Tibetan Plateau. On the basis of both peat and
pollen data from our study site, we argue that the mid-Holocene
climate was warm and wet. The delayed warming and wetting, as
compared to the conventional notion of the early Holocene monsoon
maximum induced by peak summer insolation, was likely caused by
the interplays of multiple large-scale boundary conditions, including
direct and indirect insolation controls, remnant ice sheets and seasurface temperature, and sea-level change. Weakening monsoon
intensity and decreasing summer insolation were responsible for the
long-term drying and cooling climate trend since the mid-Holocene.
Temperature is an important part of Holocene climate change in the
eastern Tibetan Plateau, despite previous emphasis on change of
monsoon-derived precipitation in driving vegetation and environmental changes.
Acknowledgments
We thank Za Dang, Shuo Chen, Jiaju Zhao for field coring assistance; Ulrike Herzschuh, Frank Schlütz and an anonymous reviewer
for helpful and constructive comments; and the University of California Irvine Keck AMS Laboratory for 14C dating analysis. This
research was supported by National Basic Research Program of
China (973 Program, grant #2010CB950202), the National Natural
Science Foundation of China (NSFC Grants #41071126 and
#41021091) and the US National Science Foundation (NSF ATM
0628455).
References
Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 million
years. Quaternary Science Reviews 10, 297e317.
1183
Camoin, G.F., Colonna, M., Montaggioni, L.F., Casanova, J., Faure, G., Thomassin, B.A.,
1997. Holocene sea level changes and reef development in the southwestern
Indian Ocean. Coral Reefs 16, 247e259.
Chen, F.H., Bloemendal, J., Zhang, P.Z., Liu, G.X., 1999. An 800 ky proxy record of
climate from lake sediments of the Zoige Basin, the Eastern Tibetan Plateau.
Palaeogeography, Palaeoclimatology, Palaeoecology 151, 307e320.
Cheng, B., 2006. Late glacial and Holocene palaeovegetation and palaeoenvironment changes in the Gonghe Basin, Tibetan Plateau. PhD dissertation, Lanzhou
University, Lanzhou, China.
Cheng, H., Edwards, R.L., Broecker, W.S., Denton, G.H., Kong, X.G., Wang, Y.J.,
Zhang, R., Wang, X.F., 2009. Ice age terminations. Science 326, 248e252.
Dean, W.E., 1974. Determination of carbonate and organic matter in calcareous
sediments and sedimentary rocks by loss on ignition: comparison with other
methods. Journal of Sedimentary Petrology 44, 242e248.
Dykoski, C.A., Edwards, R.L., Cheng, H., Yuan, D.X., Cai, Y.J., Zhang, M.L., Lin, Y.S.,
Qing, J.M., An, Z.S., Revenaugh, J., 2005. A high-resolution, absolute-dated
Holocene and deglacial Asian monsoon record from Dongge Cave, China. Earth
and Planetary Science Letters 233, 71e86.
Fægri, K., Iversen, J., 1989. Textbook of Pollen Analysis, fourth ed. John Wiley and
Sons, London, UK.
Griffiths, M.L., Drysdale, R.N., Gagan, M.K., Zhao, J.X., Ayliffe, L.K., Hellstrom, J.C.,
Hantoro, W.S., Frisia, S., Feng, Y.X., Cartwright, I., St. Pierre, E., Fischer, M.J.,
Suwargadi, B.W., 2009. Increasing AustralianeIndonesian monsoon rainfall
linked to early Holocene sea-level rise. Nature Geoscience 2, 636e639.
Grimm, E.C., 1987. CONISS: a Fortran 77 p.ogram for stratigraphically constrained
cluster analysis by the method of incremental sum of squares. Computers &
Geosciences 13, 13e35.
Herzschuh, U., 2006. Palaeo-moisture evolution in monsoonal Central Asia during
the last 50,000 years. Quaternary Science Reviews 25, 163e178.
Herzschuh, U., Winter, K., Wuennemann, B., Li, S.J., 2006. A general cooling trend on
the central Tibetan Plateau throughout the Holocene recorded by the Lake
Zigetang pollen spectra. Quaternary International 154/155, 113e121.
Herzschuh, U., Birks, H.J.B., 2010. Evaluating the indicator value of Tibetan pollen
taxa for modern vegetation and climate. Reviews of Palaeobotany and Palynology 160, 197e208.
Herzschuh, U., Birks, H.J.B., Liu, X.Q., Kubatzki, C., Lohmann, G., 2010. What caused
the mid-Holocene forest decline on the TibeteQinghai Plateau. Global Ecology
and Biogeography 19, 278e286.
Hong, Y.T., Hong, B., Lin, Q.H., Zhu, Y.X., Shibata, Y., Hirota, M., Uchida, M., Leng, X.T.,
Jiang, H.B., Xu, H., Yi, L., 2003. Correlation between Indian Ocean summer
monsoon and North Atlantic climate change during the Holocene. Earth and
Planetary Science Letters 211, 371e380.
Hou, X.Y. (Ed.), 2001. Vegetation Atlas of China (Scale: 1:1,000,000): Map I-48.
Science Press, Beijing.
Hu, C.Y., Henderson, G.M., Huang, J.H., Xie, S.C., Sun, Y., Johnson, K.R., 2008.
Quantification of Holocene Asian monsoon rainfall from spatially separated
cave records. Earth and Planetary Science Letters 266, 221e232.
Institute of Geography, 1990. Map of the QinghaieTibet Plateau. Science Press,
Beijing (in Chinese), pp. 102e112.
Jiang, W.Y., Gao, Z.T., Sun, X.J., Wu, H.B., Chu, G.Q., Yuan, B.Y., Hatté, C., Guiot, J.,
2006. Reconstruction of climate and vegetation changes of the Lake Bayanchagan (Inner Mongolia): Holocene variability of the East Asian monsoon.
Quaternary Research 65, 411e420.
Jones, M.C., Yu, Z.C., 2010. Rapid deglacial and early Holocene expansion of peatlands in Alaska. Proceedings of National Academy of Sciences USA 107,
7347e7352.
Joosten, H., Haberl, A., Schumann, M., 2008. Degradation and restoration of peatlands on the Tibetan Plateau. Peatlands International 1, 31e35.
Kaplan, M.R., Wolfe, A.P., 2006. Spatial and temporal variability of Holocene
temperature in the North Atlantic region. Quaternary Research 65, 223e231.
Kaufman, D.S., Ager, T.A., Anderson, N.J., Anderson, P.M., Andrews, J.T., Bartlein, P.J.,
Brubaker, L.B., Coats, L.L., Cwynar, L.C., Duvall, M.L., Dyke, A.S., Edwards, M.E.,
Eisner, W.R., Gajewski, K., Geirsdóttir, A., Hu, F.S., Jennings, A.E., Kaplan, M.R.,
Kerwin, M.W., Lozhkin, A.V., MacDonald, G.M., Miller, G.H., Mock, C.J.,
Oswald, W.W., Otto-Bliesner, B.L., Porinchu, D.F., Rühland, K., Smol, J.P.,
Steig, E.J., Wolfe, B.B., 2004. Holocene thermal maximum in the western Arctic
(0e180 W). Quaternary Science Reviews 23, 529e560.
Kutzbach, J.E., 1981. Monsoon climate of the early Holocene: climate experiment using the earth’s orbital parameters for 9000 years ago. Science 214,
59e61.
Liu, H.Y., Cui, H.T., Pott, R., Speier, M., 1999. The surface pollen of the woodlandsteppe ecotone in southeastern Inner Mongolia, China. Review of Palaeobotany
and Palynology 105, 237e250.
Lu, H.Y., Wu, N.Q., Yang, X.D., Shen, C.M., Zhu, L.P., Wang, L., Li, Q., Xu, D.K.,
Tong, G.B., Sun, X.J., 2008. Spatial pattern of Abies and Picea surface pollen
distribution along the elevation gradient in the QinghaieTibetan Plateau and
Xinjiang, China. Boreas 37, 254e262.
MacDonald, G.M., Velichko, A.A., Kremenetski, C.V., Borisova, O.K., Goleva, A.A.,
Andreev, A.A., Cwynar, L.C., Riding, R.T., Forman, S.L., Edwards, T.W.D.,
Aravena, R., Hammarlund, D., Szeicz, J.M., Gattaulin, V.N., 2000. Holocene
treeline history and climate change across northern Eurasia. Quaternary
Research 53, 302e311.
MacDonald, G.M., Beilman, D.W., Kremenetski, K.V., Sheng, Y., Smith, L.C.,
Velichko, A.A., 2006. Rapid development of the circumarctic peatland complex
and atmospheric CH4 and CO2 variations. Science 314, 285e288.
1184
Y. Zhao et al. / Quaternary Science Reviews 30 (2011) 1173e1184
Miehe, G., Miehe, S., Kaiser, K., Reudenbach, C., Behrendes, L., Duo, L., Schlütz, F., 2009.
How old is pastoralism in Tibet? An ecological approach to the making of a Tibetan
landscape. Palaeogeography, Palaeoclimatology, Palaeoecology 276, 130e147.
Overpeck, J., Anderson, D., Trumbore, S., Prell, W., 1996. The Southwest Indian
Monsoon over the last 18000 years. Climate Dynamics 12, 213e225.
Peltier, W.R., Fairbanks, R.G., 2006. Global glacial ice volume and Last Glacial
Maximum duration from an extended Barbados sea level record. Quaternary
Science Reviews 25, 3322e3337.
Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Bertrand, C.J.H.,
Blackwell, P.G., Buck, C.E., Burr, G.S., Cutler, K.B., Damon, P.E., Edwards, R.L.,
Fairbanks, R.G., Friedrich, M., Guilderson, T.P., Hogg, A.G., Hughen, K.A.,
Kromer, B., McCormac, F.G., Manning, S.W., Ramsey, C.B., Reimer, R.W.,
Remmele, S., Southon, J.R., Stuiver, M., Talamo, S., Taylor, F.W., van der Plicht, J.,
2004. Intcal04 terrestrial radiocarbon age calibration, 0e26 cal kys BP. Radiocarbon 46, 1029e1058.
Rimbu, N., Lohmann, G., Lorenz, S.J., Kim, J.H., Schneider, R.R., 2004. Holocene climate
variability as derived from alkenone sea surface temperature and coupled
oceaneatmosphere model experiments. Climate Dynamics 23, 215e227.
Ruddiman, W.F., 2008. Earth’s Climate: Past and Future, second ed. W.H. Freeman
and Company, New York, pp. 138e142.
Schlütz, F., Lehmkuhl, F., 2009. Holocene climatic change and the nomadic
Anthropocene in eastern Tibet: palynological and geomorphological results
from the Nianbaoyeze Mountains. Quaternary Science Reviews 28, 1449e1471.
Shao, X.H., Wang, Y.J., Cheng, H., Kong, X.G., Wu, J.Y., 2006. Long-term trend and
abrupt events of the Holocene Asian monsoon inferred from a stalagmite d18O
record from Shennongjia in Central China. Chinese Science Bulletin 51, 80e86.
Shen, C.M., 2003. Millennial-scale variations and centennial-scale events in the
Southwest Asian monsoon: pollen evidence from Tibet. PhD dissertation,
Louisana State University, Banta Rouge.
Shen, C.M., Tang, L.Y., 1996. Vegetation and climate during the last 22 000 years in
Zoige Basin. Acta Micropalaeontologia Sinica 13, 401e406.
Shen, C.M., Liu, K.-B., Tang, L.Y., Overpeck, J.T., 2006. Quantitative relationships
between modern pollen rain and climate in the Tibetan Plateau. Review of
Palaeobotany and Palynology 140, 61e77.
Shen, J., Liu, X.Q., Wang, S.M., Matsumoto, R., 2005. Palaeoclimatic changes in the Qinghai
Lake area during the last 18,000 years. Quaternary International 136, 131e140.
Sugita, S., 1993. A model of pollen source area for an entire lake surface. Quaternary
Research 39, 239e244.
Thelaus, M., 1992. Some characteristics of the mire development in Hongyuan
County, eastern Tibetan Plateau. In: International Peat Congress Proceedings,
pp. 334e351.
Vitt, D.H., 2006. Functional characteristics and indicators of boreal peatlands. In:
Wieder, R.K., Vitt, D.H. (Eds.), Boreal Peatland Ecosystems Ecological Studies,
vol. 188. Springer, pp. 9e24.
Vitt, D.H., Halsey, L.A., Bauer, J.E., Campbell, C., 2000. Spatial and temporal trends of
carbon sequestration in peatlands of continental western Canada through the
Holocene. Canadian Journal of Earth Science 37, 683e693.
Wang, Y.J., Cheng, H., Edwards, R.L., He, Y.Q., Kong, X.G., An, Z.S., Wu, J.Y., Kelly, M.J.,
Dykoski, C.A., Li, X.D., 2005. The Holocene Asian monsoon: links to solar
changes and North Atlantic climate. Science 308, 854e857.
Wang, Y.J., Cheng, H., Edwards, R.L., Kong, X.G., Shao, X.H., Chen, S.T., Wu, J.Y.,
Jiang, X.Y., Wang, X.F., An, Z.S., 2008. Millennial- and orbital-scale changes in
the East Asian monsoon over the past 224,000 years. Nature 451, 1090e1093.
Xiao, J.L., Xu, Q.H., Nakamura, T., Yang, X.L., Liang, W.D., Inouchi, Y., 2004. Holocene
vegetation variation in the Daihai Lake region of north-central China: a direct
indication of the Asian monsoon climatic history. Quaternary Science Reviews
23, 1669e1679.
Yan, G., Wang, F.B., Shi, G.R., Li, S.F., 1999. Palynological and stable isotopic study of
palaeoenvironmental changes on the northeastern Tibetan Plateau in the last
30,000 years. Palaeogeography, Palaeoclimatology, Palaeoecology 153, 147e159.
Yang, X., Ma, N., Dong, J., Zhu, B., Xu, B., Ma, Z., Liu, J., 2010. Holocene hydrological
and climatic changes in the Badain Jaran Desert, western China. Quaternary
Research 73, 10e19.
Yang, X., Scuderi, L., 2010. Hydrological and climatic changes in deserts of China
since the Late Pleistocene. Quaternary Research 73, 1e9.
Yu, G., Tang, L.Y., Yang, X.D., Ke, X.K., Harrison, S.P., 2001. Modern pollen samples
from alpine vegetation on the Tibetan Plateau. Global Ecology & Biogeography
10, 503e519.
Yu, Z.C., Beilman, D.W., Jones, M.C., 2009. Sensitivity of northern peatlands to
Holocene climate change. In: Baird, A., Belyea, L., Comas, X., Reeve, A., Slater, L.
(Eds.), AGU Geophysical Monograph. Carbon Cycling in Northern Peatlands, vol.
184, pp. 55e69.
Yu, Z.C., Campbell, I.D., Campbell, C., Vitt, D.H., Bond, G.C., Apps, M.J., 2003. Carbon
sequestration in western Canadian peat highly sensitive to Holocene wetedry
climate cycles at millennial time scales. The Holocene 13, 801e803.
Yu, Z.C., Loisel, J., Brosseau, D.P., Beilman, D.W., Hunt, S.J., 2010. Global peatland
dynamics since the Last Glacial Maximum. Geophysical Research Letters 37,
L13402. doi:10.1029/2010GL043584.
Zhou, W.J., Lu, X.F., Wu, Z.K., et al., 2002. Peat record reflecting Holocene climatic
change in the Zoige Plateau and AMS radiocarbon dating. Chinese Science
Bulletin 47, 66e70.
Zhou, W.J., Yu, S.-Y., Georges, B., Kukla, G.J., Jull, A.J.T., Xian, F., Xiao, J.Y.,
Colman, S.M., Yu, H.G., Liu, Z., Kong, X.H., 2010. Postglacial changes in the Asian
summer monsoon system: a pollen record from the eastern margin of the
Tibetan Plateau. Boreas 39, 528e539.
Zhou, W.J., Yu, X.F., Timothy Jull, A.J., Burr, G., Xiao, J.Y., Lu, X.F., Xian, F., 2004. Highresolution evidence from southern China of an early Holocene optimum and a midHolocene dry event during the past 18,000 years. Quaternary Research 62, 39e48.