The 9.2 ka event in Asian summer monsoon area: the strongest millennial scale collapse of the monsoon during the Holocene

Abstract

Numerous Holocene paleo-proxy records exhibit a series of centennial-millennial scale rapid climatic events. Unlike the widely acknowledged 8.2 ka climate anomaly, the likelihood of a significant climate excursion at around 9.2 cal ka BP, which has been notably recognized in some studies, remains to be fully clarified in terms of its magnitude and intensity, as well as its characteristics and spatial distributions in a range of paleoclimatic records. In this study, a peat sediment profile from the Dajiuhu Basin in central China was collected with several geochemical proxies and a pollen analysis carried out to help improve understanding of the climate changes around 9.2 cal ka BP. The results show that the peat development was interrupted abruptly at around 9.2 cal ka BP, when the chemical weathering strength decreased and the tree-pollen declined. This suggests that a strong drier regional climatic event occurred at around 9.2 cal ka BP in central China, which was, in turn, probably connected to the rapid 9.2 ka climate event co-developing worldwide. In addition, based on the synthesis of our peat records and the other Holocene hydrological records from Asian summer monsoon (ASM) region, we further found that the 9.2 ka event probably constituted the strongest abrupt collapse of the Asian monsoon system during the full Holocene interval. The correlations between ASM and the atmospheric ¹⁴C production rate, the North Atlantic drift ice records and Greenland temperature indicated that the weakened ASM event at around 9.2 cal ka BP could be interpreted by the co-influence of external and internal factors, related to the changes of the solar activity and the Atlantic Meridional Overturning Circulation (AMOC).

Full text

Vol.:(0123456789)
1 3
Clim Dyn
DOI 10.1007/s00382-017-3770-2
The 9.2ka event inAsian summer monsoon area: thestrongest
millennial scale collapse ofthemonsoon duringtheHolocene
WenchaoZhang1,2,3· HongYan1,2 · JohnDodson1,4· PengCheng1·
ChengchengLiu1,3· JianyongLi1· FengyanLu1· WeijianZhou1,2· ZhishengAn1,2
Received: 21 December 2016 / Accepted: 12 June 2017
© Springer-Verlag GmbH Germany 2017
the strongest abrupt collapse of the Asian monsoon system
during the full Holocene interval. The correlations between
ASM and the atmospheric 14C production rate, the North
Atlantic drift ice records and Greenland temperature indi-
cated that the weakened ASM event at around 9.2cal ka
BP could be interpreted by the co-influence of external and
internal factors, related to the changes of the solar activ-
ity and the Atlantic Meridional Overturning Circulation
(AMOC).
Keywords Dajiuhu peat· Central China· Paleoclimate
records· Abrupt climate changes· 9.2ka BP event· Weak
Asian summer monsoon
1 Introduction
A large number of Holocene climate reconstructions reveal
a climatic pattern far from stable throughout the Holocene.
The most prominent feature is a series of decadal-to-cen-
tennial scale oscillations and/or abrupt events (Andrews
et al. 1997; Axford et al. 2009; Bügelmayer-Blaschek
etal. 2016; Bond etal. 2001, 1997; Chen etal. 2015; Dixit
etal. 2014; Fleitmann etal. 2003, 2008; Hou etal. 2012;
Isono etal. 2009; Liu et al. 2014; Mayewski etal. 2004;
Moros etal. 2006; Owen etal. 2016; Schulz and Paul 2002;
Shao etal. 2006; Soon etal. 2014; Wang etal. 2005; Wan-
ner etal. 2011; Zhou etal. 2007). These previous studies
show that there are at least eight large and rapid climatic
events, not counting the Little Ice Age (LIA), recorded in
the North Atlantic drift ice records throughout the Holo-
cene, including events at 8.2, 9.4, 10.3 and 11.1ka in the
early Holocene (Bond etal. 2001, 1997). The δ18O and ice
accumulation signals of Greenland ice cores, along with
some lake, stalagmite and tree ring records at northern
Abstract Numerous Holocene paleo-proxy records
exhibit a series of centennial-millennial scale rapid climatic
events. Unlike the widely acknowledged 8.2 ka climate
anomaly, the likelihood of a significant climate excursion
at around 9.2cal ka BP, which has been notably recognized
in some studies, remains to be fully clarified in terms of its
magnitude and intensity, as well as its characteristics and
spatial distributions in a range of paleoclimatic records. In
this study, a peat sediment profile from the Dajiuhu Basin
in central China was collected with several geochemical
proxies and a pollen analysis carried out to help improve
understanding of the climate changes around 9.2cal ka BP.
The results show that the peat development was interrupted
abruptly at around 9.2cal ka BP, when the chemical weath-
ering strength decreased and the tree-pollen declined. This
suggests that a strong drier regional climatic event occurred
at around 9.2 cal ka BP in central China, which was, in
turn, probably connected to the rapid 9.2ka climate event
co-developing worldwide. In addition, based on the synthe-
sis of our peat records and the other Holocene hydrologi-
cal records from Asian summer monsoon (ASM) region,
we further found that the 9.2ka event probably constituted
* Hong Yan
yanhong@ieecas.cn
1 State Key Laboratory ofLoess andQuaternary Geology,
Institute ofEarth Environment, Chinese Academy
ofSciences, Xi’an710061, China
2 Interdisciplinary Research Center ofEarth Science Frontier
(IRCESF) andJoint Center forGlobal Change Studies
(JCGCS), Beijing Normal Universtity, Beijing100875, China
3 University ofChinese Academy ofSciences, Beijing100049,
China
4 School ofEarth andEnvironmental Sciences, University
ofWollongong, Wollongong, NSW2500, Australia

W.Zhang et al.
1 3
high latitudes, also highlight cold and dry climate oscilla-
tions during the early Holocene (Hu etal. 2003; Korhola
etal. 2002; Mcdermott etal. 2001; Rasmussen etal. 2007;
Spurk etal. 2002). Among these early Holocene millennial-
scale climate oscillations, the 8.2ka climate anomaly event
has received much more attention and has been rather well
studied. It has been suggested that this rapid climate cool-
ing event occurs robustly and consistently in many parts of
the world, such as Greenland, North Atlantic, Mediterra-
nean, Southwestern Asia, East Africa, North America and
Antarctica (Alley and Ágústsdóttir 2005; Alley etal. 1997;
Barber et al. 1999; Fleitmann et al. 2007, 2003; Johnsen
etal. 2001; Pross etal. 2009; Stager and Mayewski 1997).
Quite a few hydrological records from the Asian summer
monsoon (ASM) area, including lake sediment, speleo-
them, marine sediment and ice core records, also record
the 8.2ka climatic event (An etal. 2012; Chen etal. 2001;
Dykoski et al. 2005; Fleitmann et al. 2003; Gupta et al.
2003, 2005; Hu etal. 2008; Jin etal. 2003, 2004; Neff etal.
2001; Wang etal. 2002; Wu etal. 1995).
In contrast to the 8.2ka event, other abrupt climate oscil-
lations during the early Holocene, such as the 9.2ka event,
have attracted much less attention, and the spatial–tempo-
ral changes of the ASM around the 9.2ka event are still
not well established. Based on four well-dated stalagmite
δ18O records, Fleitmann etal. (2008) point out that a dry-
ing occurs in the Northern tropics around 9.2 cal ka BP,
which is similar to the 8.2ka event. However, the number
of proxy records and the areal extent in that study are still
insufficient and thus the characteristics and dynamics of
the 9.2ka climate event, particularly the comparison of 9.2
versus 8.2ka events, over the broader areas affected by the
ASM are still unclear. For example, the signal of the 8.2ka
event is stronger than the 9.2ka event in the Greenland ice
record (Rasmussen etal. 2006; Vinther etal. 2006), while
the 9.2 ka event is stronger in some hydrological records
from the ASM area (An etal. 2012; Dykoski etal. 2005;
Gupta etal. 2003, 2005; Jia etal. 2015). Thus the strength
of the 9.2 ka event versus 8.2 ka event within the ASM
area, as well as the relative role of high latitude versus low
latitude driving these two events, needs further detailed
examination.
Peatlands are one of the most important paleoclimate
archives, and play an important role in the reconstruction
of precipitation, vegetation and other paleoclimate informa-
tion (Anderson 1998; Chambers etal. 1997; Chambers and
Charman 2004; Horák-Terra etal. 2015; Xue etal. 2015).
For example, lots of geochemical proxies in peat sediment,
such as elements, stable isotope and organic matters, as
well as pollen records, have already been generated reli-
ably to understand the paleo-hydrological and temperature
changes (Burrows etal. 2016; Ferrat etal. 2012; He etal.
2015; Knaap etal. 2011; Kylander etal. 2013).
Dajiuhu is a rare alpine peatland in the subtropical
region of China and has formed a thick continuous peat
stratum since the last deglaciation period under the domi-
nant control of the East Asian summer monsoon (EASM).
Therefore the Dajiuhu peat can potentially improve our
knowledge of the EASM intensity during the early Holo-
cene in central China, which includes the time interval
around 9.2cal ka BP.
In this study, we collected a sediment core profile (Y1)
from the Dajiuhu Basin, and measured several paleocli-
mate proxies, including total organic carbon (TOC), total
nitrogen (TN), Al, Ti, Rb/Sr, δ13C and pollen analysis, to
investigate the peat development, mineral inputs, chemical
weathering and vegetation composition around 9.2cal ka
BP, respectively (Jin etal. 2006; Koinig etal. 2003; Li etal.
2013; O’Leary 1988). We also compared our Dajiuhu peat
records with other available hydrological records in the
ASM area in order to provide an improved synthesis and
understanding of the ASM variations at around 9.2cal ka
BP.
2 Materials andmethods
2.1 Study site andsampling
The Dajiuhu Basin (N31°24′–N31°33′,
E109°56′–E110°11′) is a closed karst basin with an inde-
pendent watershed, located at 1700–1760 m a.s.l. in the
west of the Shennongjia Mountains, in Hubei Province,
central China (Fig.1a). The basin topography is flat, and
surrounded by steep mountains of dolomite and dolomitic
limestone. With an area of about 16km2, the basin is cov-
ered with thick peat, and is a subalpine peatland in the
mid-latitude subtropical region. The regional climate has
an unusual seasonal variability under the strong control of
the EASM, which is marked by short, warm and wet sum-
mers and long, cold and dry winters. The average annual
temperature is 7.2 °C with the highest monthly average val-
ues of 17.1 °C in July and the lowest values of −2.4 °C in
January. The average annual precipitation is approximately
1500 mm with the maximum up to 3000 mm. Yearly
rainy days extend up to 150–200 days, mainly from April
to October, and with the humidity over 80% (Ma et al.
2009). Due to relatively high altitude, vegetation in the
Dajiuhu area has distinct communities and vertical zones.
In the Dajiuhu peat, the vegetation community is mainly
composed of swamp plants, such as species of Carex and
Sphagnum (Li etal. 2013).
Our profile was from a digged pit, and we collected
165 continuous samples through a depth sequence of
205 cm at the center of Dajiuhu Basin (31°28′48.93″N,
110°00′6.65″E, Fig. 1b), where there was minimal

The 9.2ka event inAsian summer monsoon area: thestrongest millennial scale collapse ofthe…
1 3
disturbance by anthropogenic activities. The samples
were collected at 1.2cm intervals from 0 to 188cm depth,
then at 2 cm intervals from the depth of 188 to 198cm,
and at about 2.3cm for the intervals from 198 to 205cm
(Zhang et al. 2016). The profile (Y1) was described and
photographed in the field: The upper part of the sediment
(0-177cm) was mainly composed of dark peat with plant
macro-remains. The lower part (177–205cm) was largely
grey clay with light humification (Fig.2).
2.2 Experiments
To obtain a high-resolution chronology for the sequence,
ten bulk samples and two extractive pollen samples were
taken for radiocarbon dating. In order to extract pure pol-
len, carbonate was removed from the peat samples using
diluted hydrochloric acid, then NaOH and NaClO2 was
added and sieved to remove coarse organic matter. After
washing with deionized water and drying, two pollen resi-
due samples, as well as ten bulk samples treated by the
acid-base-acid chemical technique, were measured by the
Accelerator Mass Spectrometry (AMS) in the Institute of
Earth Environment, Chinese Academy of Sciences, Xi’an.
The AMS 14C ages were calibrated into 2σ calendar ages
and we obtained 12 median calibrated ages using the
IntCal13 calibration curve (Reimer etal. 2013).
After drying for 48 h at 50 °C, samples for geochemi-
cal analysis were ground to homogeneity while coarser
samples were removed by sieving through a 150μm mesh
filter. For TOC and TN analyses, we added hydrochloric
acid (5%) to the grey clay samples and mixed these until
all evidence of carbonate was removed. Then the pH of
these samples was adjusted to pH 7 with deionized water.
These samples, as well as the dark peat samples without
Fig. 1 Location of the Dajiuhu Basin and the sampling site. a The
colour shading on the map indicates annual mean precipitation as
derived from NCEP reanalysis 2 from January 1979 to December
2010. White dots show the locations of the climate proxy records in
the ASM area mentioned in this paper (An etal. 2012; Chen et al.
2015; Dong etal. 2010; Dykoski etal. 2005; Fleitmann etal. 2003;
Gupta etal. 2003, 2005; Hu et al. 2008; Jia et al. 2015; Neff et al.
2001; Zhou etal. 2007). The red dot shows the location of the Daji-
uhu peat. b The topography of Dajiuhu Basin. The sampling site (Y1)
is located in the middle of the Dajiuhu peatland, marked by red dot.
And other TOC/POC records mentioned in this paper are marked by
white dots. c The landscape near the sample site in January

W.Zhang et al.
1 3
pretreatment, were placed into an elemental analysis instru-
ment to determine TOC and TN after dry combustion
(ISO 1995). Several control standard samples were also
measured and their deviations of repetitions were <0.02%
(TOC) and <0.002% (TN). For chemical element analysis,
about 0.1g of each sample was taken, and weighed. Al, Ti
and Rb/Sr were determined using the Inductively Coupled
Plasma-Atomic Emission Spectrometry (ICP-AES) with
method 200.7 (EPA 2001) after digestion by HF-HNO3-
HClO4. The detection limits of Al, and Ti were 0.01 and
0.001mg/g, meeting all the high standards of accuracy and
precision. These measurements were conducted in the Lake
Sediment and Environment Laboratory of Nanjing Institute
of Geography and Limnology, Chinese Academy of Sci-
ences, Nanjing. For stable carbon isotope in the organic
matter, 76 samples (No. 43-118, 56.6–146.6 cm) were
determined using EA-Isoprime100, coupled through a con-
tinuous flow system to an isotope ratio mass spectrometer
in the Environmental Stable Isotope Laboratory of Institute
of Environment and Sustainable Development in Agricul-
ture, Chinese Academy of Agricultural Sciences, Beijing.
δ13C values were reported as per mil (‰) relative to the
Vienna Pee Dee Belemnite (VPDB) international stand-
ard. Precision estimated from repeated analyses of internal
standards was better than ±0.2‰.
The samples for pollen analysis consisted of 1.2cm con-
tiguous samples between 98.4 and 122.4cm depth from the
same core used in the geochemical analysis. They were pre-
pared in a clean pollen preparation laboratory in the Insti-
tute of Earth Environment, Chinese Academy of Sciences,
Xi’an. The samples were treated with the standard pollen
preparation procedures of spiking with Lycopodium spores,
boiling for about 5min in 5% KOH and sieving through a
120μm mesh, then acetolysis and residues were mounted
on slides in glycerol (Moore etal. 1991). Pollen and spores
were then identified and counted until a minimum of about
400–500 grains had been observed.
3 Results
3.1 Radiocarbon ages andsedimentation rates
Table 1 shows the result of radiocarbon age determina-
tions. In order to test the accuracy of dating results using
the bulk sediment, two samples (Y1-82 and Y1-65) were
Fig. 2 Lithological characteristics and age-depth model of peat pro-
file. The lithological characteristics of the profile are divided into
black peat (0–177cm) and grey clay sediment (177–205cm) from the
surface to the base
Table 1 AMS14C dating results of the Dajiuhu peat
Sample no. Depth (cm) Material 14C age (year BP) Error (±year) Calibrated age (cal
year BP-2σ range)
Median calibrated
age (cal year BP-2σ)
Corrected age
(cal year BP)
Y1-150 18-19.2 Peat 2010 29 1887–2039 1959 2265
Y1-150 18-19.2 Peat 2021 31 1891–2058 1970
Y1-134 37.2–38.4 Peat 3828 26 4101–4400 4219 4519
Y1-116 58.8–60 Peat 5250 27 5927–6176 5993 6293
Y1-99 79.2–80.4 Peat 7091 31 7850–7972 7930 8230
Y1-82 99.6-100.8 Pollen 7710 36 8416–8576 8491 8491
Y1-82 99.6-100.8 Peat 7385 35 8056–8329 8223
Y1-65 120-121.2 Pollen 8818 35 9696–10,145 9857 9857
Y1-65 120-121.2 Peat 8545 37 9484–9549 9526
Y1-47 141.6-142.8 Peat 9261 40 10,291–10,562 10,442 10,742
Y1-31 160.8–162 Peat 9988 48 11,263–11,702 11,455 11,755
Y1-14 181.2-182.4 Organic matter 11,230 45 13,018–13,188 13,096 13,096
Y1-3 198-200.3 Organic matter 12,039 61 13,750–14,061 13,891 13,891

The 9.2ka event inAsian summer monsoon area: thestrongest millennial scale collapse ofthe…
1 3
taken for radiocarbon dating using both the bulk sediment
and pollen grains. The results show that the median cali-
brated ages from bulk peat were ~300 years younger than
those from pollen, which could be connected with some
younger plant roots penetrating into the peat. Thus, the dat-
ing results from black peat were added with 300years in
order to remove the “New Carbon” effect (last column in
Table1). A linear interpolation between two corrected ages
was used to establish an age-depth model with the sediment
surface assumed to be zero age (Fig.2). The basal age was
about 14.1cal ka BP, which was estimated by extrapola-
tion. According to the age-depth model, the sedimenta-
tion rates between any two age-control points showed little
changes, with a mean of 0.015cm/yr except the fastest rate
(0.078cm/year) between 8.5 and 8.2cal ka BP.
3.2 TOC andTN profiles
The TOC and TN contents showed relatively similar and
stable variation during 14.1–2.0cal ka BP (Fig.3), except
for two short stages (14.1–11.5cal ka BP and 10.5–9.0cal
ka BP). The TOC contents had variable values from 0.77
to 51.77% and the TN contents oscillated between 0.118
and 2.321%. The lowest values for the whole sequence
appeared during 10.5–9.0 cal ka BP. Both TOC and TN
values decreased rapidly from about 10.5cal ka BP, and
although they increased slightly during 10.0–9.7cal ka BP,
the values then fell most sharply from around 9.7cal ka
BP and reached their lowest value at about 9.3cal ka BP.
Within a short time interval (9.7–9.3cal ka BP), the TOC
values fell to 23.50% from a high level of 38.43%, and the
TN values fell from 1.702% down to 1.095%.
3.3 Al, Ti andRb/Sr profiles
Al and Ti elemental concentrations showed a similar pat-
tern of changes between 14.1 and 2.0cal ka BP, and they
were negatively correlated with the TOC and TN contents
(Fig.3). The values of Al and Ti ranged from 8.88 to 87.50
and 0.72 to 4.80mg/g, respectively. Both Al and Ti values
were high during the 10.5–9.0cal ka BP interval (Fig.3).
Elemental Al and Ti concentrations increased from about
10.5 cal ka BP, and reached a minor peak near 10.0 cal
ka BP. After a decreasing trend from 10.0 to 9.7 cal ka
BP, they all increased sharply again within 400 years
(9.7–9.3cal ka BP) (Fig.3). Likewise, the Rb/Sr ratio val-
ues took on a similar pattern, where the period of highest
values were constant at 10.5–9.0cal ka BP, and minor peak
values (2.16 and 2.25mg/g) occurred at around 10.0 and
9.2cal ka BP with a dip during 9.7–9.4cal ka BP (Fig.3).
In addition, the strong fluctuations were also observed in
Al, Ti and Rb/Sr records at around 8.2cal ka BP.
3.4 Pollen diagram
Figure4 shows a pollen diagram of selected taxa calculated
against a sum of total pollen grains excluding Cyperaceae
and aquatic taxa. Rare types, generally occurring in only a
few samples and with values below 2% are not shown.
The Constrained Incremental Sum of Squares (CONISS)
procedure was used to yield a zonation of the diagram.
Four major zones from I to IV (youngest to oldest) were
recognized and shown on the diagram (Fig.4).
3.4.1 Zone IV (9.9–9.6cal ka BP):
The lowermost zone in the diagram had relatively low pol-
len concentration and the most common tree taxa were
Quercus, Fagus, Carpinus, Betula, Ulmus and Pinus.
Euphorbiaceae and Castanopsis-Castanea were also pre-
sent in this zone but they declined to low levels. Myrtaceae
values were also relatively low. Poaceae values reached
about 10% but declined toward the end of the Zone and
Cyperaceae were above about 25% of the total pollen sum.
The vegetation was likely a mixed temperate forest with
low values of shrubby taxa. There was a diverse array of
herbaceous taxa including Poaceae suggesting open areas,
perhaps around or on the peatland. The occurrence of Pota-
mogeton and Sphagnum with Geraniaceae and some fern
spores and the abundance of Cyperaceae suggested a series
of hummocks and hollows on the peatland.
Fig. 3 Characteristics of TOC, TN, Rb/Sr, Al and Ti profiles in the
Dajiuhu peat records. The blue bars indicate the 8.2, 9.2ka and YD
events, respectively

W.Zhang et al.
1 3
3.4.2 Zone III (9.6–9.1cal ka BP):
Zone III had generally higher values of Quercus, Juglans
and Castanea-Castanopsis values than Zone IV, while
Pinus, Ulmus and Betula values were similar. Poaceae
values were lower than those in Zone IV, but showed an
increasing pattern toward the end of the period, while
Cyperaceae values were generally higher. Sphagnum and
Potamogeton values remained similar, but Sphagnum
decreased at the end of the Zone. Overall it seemed that the
forest composition was similar but began to open up by the
end of Zone III.
Fig. 4 Simplified pollen
percentage diagram of Dajiuhu
peat

The 9.2ka event inAsian summer monsoon area: thestrongest millennial scale collapse ofthe…
1 3
3.4.3 Zone II (9.1–8.6cal ka BP):
The largest break in similarity occurred between Zones III
and II. While Quercus, Castanea-Castanopsis and Pinus
values were similar, Fagus, Betula, and to a lesser degree
Carpinus values were lower. Salix values showed a peak
in the middle of the Zone and Poaceae values reached
a maxima in the whole record peaking above 40% of the
pollen sum but they declined toward the end of the Zone.
Myrtaceae pollen values were low to nearly zero. Cyper-
aceae values showed a decline across the Zone as do
Sphagnum spores. Monolete spores showed a maximum,
which suggested that the forest canopy had opened up com-
pared to previous zones. Near the top of the Zone several
taxa, including Apocynaceae, Araliaceae and Rubiaceae,
declined and even dropped out of the record in the follow-
ing Zone. The vegetation had a much more open aspect in
this Zone and the peatland surface was relatively drier.
3.4.4 Zone I (8.6–8.5cal ka BP):
Pollen concentration was relatively higher in this zone.
Herein lower Pinus and evergreen Quercus values and
higher Betula, deciduous Quercus, Fagus and Juglans
values were evident compared to Zone II. Corylus values
increased as did Rosaceae and Liliaceae. Poaceae values
showed an increase but were generally lower than in Zone
II. Cyperaceae values were high at over 20% of the pol-
len sum. In general there were shifts in forest composition
and understorey shrub taxa and the vegetation had a more
closed biome aspect compared to Zone II. The peatland
surface had less Sphagnum present.
3.5 δ13C profile
The δ13C values fluctuated between −28.34 and −27.78‰
during 11.0–6.1cal ka BP (Fig.5). Between 9.1 and 8.6cal
ka BP, it had more negative values, falling from −28.04 to
−28.34‰, though there was a small peak around 8.9cal ka
BP.
4 Discussion
4.1 Peat development interruption ataround9.2cal ka
BP intheDajiuhu Basin
The changes in both TOC and TN values in peat sediments
are nominally used as the proxies for the peat development,
which is mainly influenced by the initial productivity, the
input of terrigenous organic detritus and the preservative
ability after those organic matters deposited. Normally,
warm and wet climate conditions are conducive to the rapid
Fig. 5 The characteristics of geochemical and pollen records around
9.2 cal ka BP. The increased Rb/Sr ratio and decreased TOC con-
tents point to the weak chemical weathering and the interruption of
the peat development around 9.2ka respectively. The δ13C and pollen
records, indicating more grasses, less trees and aquatic plants, could
reflect the abrupt changes of ecological system in response to climate
changes around 9.2cal ka BP. A delayed response of about 200–300
years was observed in pollen and δ13C records relative to the geo-
chemical records

W.Zhang et al.
1 3
accumulation of the organic matters, due to dense vegeta-
tion cover and high input of the organic detritus, resulting
in the high TOC and TN contents in peat sediments (Chai
1990; Li etal. 2013).
In this study, the values of TOC and TN contents
increased greatly during 14.1–11.5cal ka BP (Fig.3), sug-
gesting that the Dajiuhu peat probably initiated and devel-
oped during the last deglaciation. After 11.5cal ka BP, the
end of the Younger Dryas (YD) event, the TOC contents
increased from lower than 30% to about 50% within 400
years (Fig.3). Thereafter the TOC contents were relatively
high and the Dajiuhu peatland developed at a stable pace
during the whole Holocene, except for the period between
10.5 and 9.0cal ka BP (Fig.3). The sharp decrease of TOC
and TN during 10.5–9.0cal ka BP indicated that the peat
development in our sampling site was probably interrupted,
which may be a reflection of an abrupt climate change asso-
ciated with the EASM variability.
The TOC variation in the Dajiuhu Basin has also been
investigated in several previous studies (He et al. 2015;
Huang 2009; Ma etal. 2008). Although the variation pat-
terns of these TOC/POC profiles from different sampling
sites presented apparent asynchronies since the last degla-
ciation, all of these records revealed obvious low values
around 10–9.5cal ka BP (Fig.6) (Huang 2009; Ma etal.
2008). The coherent decrease of the TOC/POC values from
different sampling sites indicated that the peat development
was interrupted across wide areas in the Dajiuhu Basin at
around 9.2cal ka BP.
4.2 Chemical weathering andclimate change
ataround9.2cal ka BP intheDajiuhu Basin
It is acknowledged that the migration ability of different
chemical elements may perform differently in the weath-
ering and erosion processes, due to differences in chemi-
cal properties and mobility. The strong contrasts in geo-
chemical behavior lead to an effective fractionation of Rb
and Sr during chemical weathering, therefore resulting in
a significant decrease in the Rb/Sr ratio in lake/peat sedi-
ments relative to the residual weathered crust when the
chemical weathering is enhanced (Jin etal. 2006). Climate
change exerts an important influence on the rate of chemi-
cal weathering (Gibbs and Kump 1994). Warm and wet
climate contributes to chemical weathering process, and
results in lower Rb/Sr ratios in the lake/peat sediments (Jin
etal. 2001).
Some chemical elements, such as Al and Ti, are usually
less mobile and stay in-situ, irrespective of the chemical
weathering conditions (Nesbitt and Young 1982). During
the periods with dry and cold climate conditions, the vege-
tation cover would decline and high amounts of dust would
be transported to the lake/peat by the means of aeolian or
soil erosion processes (Koinig et al. 2003), resulting in
increased intensity of mineral sedimentary inputs into the
lake/peat and high Al and Ti contents. Thus the changes of
Al and Ti concentrations are usually negatively correlated
with the TOC and TN in some peat sediments.
The chemical elements and Rb/Sr ratios in this
study show significant fluctuations during the period of
10.5–9.0cal ka BP, whereas the sediment lithology of the
profile (108–136cm, 9.0–10.5 cal ka BP) does not show
marked visual changes (Fig.2). Thus we assume that there
were no perturbations caused by non-climatic events, such
as geologic events or fires in our peat sediment. In addition,
the sedimentary rates during this period are stable with a
linear relationship between age and depth. This allows us
to conclude that the variations of our climate proxies reveal
natural climate-related changes around the sampling sites
during the period of 10.5-9.0cal ka BP.
During 10.5–10.0 cal ka BP, as the peat develop-
ment slowed down, both the Al and Ti concentrations
increased sharply (Fig. 3). This suggested that terrig-
enous detritus with low organic matters was transported
into the peat sediment. As well, the Rb/Sr ratios showed
a significant increase (Fig. 3), suggesting weak chemical
weathering, resulting from the cold and dry climatic con-
ditions during this period. During the period of improved
Fig. 6 The TOC/POC records at different sampling sites in the Daji-
uhu Basin. a TOC record in this study; b TOC record from Ma etal.
(2008); c POC record from Huang (2009). The blue bar indicates the
obvious decrease during 10.5–9.0cal ka BP

The 9.2ka event inAsian summer monsoon area: thestrongest millennial scale collapse ofthe…
1 3
peat development, from 10.0 to 9.7cal ka BP, Al and Ti
declined slightly and Rb/Sr ratios were low (Fig.3). This
indicated an accelerated chemical weathering process and
low aeolian deposition in the Dajiuhu Basin and a warm
and humid climate. During 9.7–9.3 cal ka BP, Al and Ti
rose significantly again, with the peak values occurring at
around 9.3cal ka BP (Fig.3), reflecting an interruption of
the peat development. Moreover, the changes of the Rb/Sr
ratios were in phase with those of the chemical elemental
contents, and increased abruptly since 9.4 cal ka BP and
reached its peak at around 9.2cal ka BP (Fig.3). It thus
seemed that the chemical weathering rate had weakened
since 9.4cal ka BP, resulting from an abrupt cold and dry
climate. We inferred that this was possibly a response to a
rapid and abnormal climate event and an abrupt weakening
of the EASM intensity.
Overall, the higher Al and Ti contents further confirmed
the interruption of the peat development in the Dajiuhu
Basin around 9.2cal ka BP, while the higher Rb/Sr ratios
indicated that the interruption was associated with the
weak chemical weathering under the possible cold/dry cli-
mate condition.
4.3 Vegetation changes intheDajiuhu Basin
andadjacentarea ataround9.2cal ka BP
The pollen diagram showed that the pollen percentages
of dominant trees varied a little, but the major change
occurred in Zone II at about 9.1cal ka BP (Fig.4). There
was a decline in tree cover, particularly deciduous taxa of
oaks, indicating an open landscape with an expansion of
some open-land taxa of ferns and grasses (Figs.4, 5). This
suggested a drier condition between 9.1 and 8.7cal ka BP.
The peatland plants, such as Cyperaceae, also had a dis-
tinct change (Fig.5), with peat development degenerating.
After this period, however, there was a decline in open-
land pollen elements, but a recovery of trees and shrubs,
including Betula, Fagus, Carpinus and deciduous Quercus,
showing that the vegetation redeveloped to a more closed
canopy biome (Fig.4). This process took over 200 years to
be completed, which could be also seen from the variations
of other pollen taxa, such as Juglans, Myrtaceae (probably
Syzygium) and Rosaceae. However, it appeared that Pinus
with low abundance locally, did not recover after the large
shift in climate (Fig.4). Our Dajiuhu records showed that
vegetation response and recovery was relatively slower
by 200–300 years in response to climate change com-
pared with the TOC, Rb/Sr and other geochemical tracers
(Fig.5). The vegetation variations were synchronous with
the δ13C values (Fig.5).
The delayed response may be because the resilience of
forest trees enables them to cope and re-adjust with cer-
tain kinds of unfavourable environmental conditions over
many years to even decades. Once a shift to an unfavour-
able environment becomes prolonged or is severe, the resil-
ience breaks down. In this case, a decline in tree cover and
an opening up of the canopy would occur in the Dajiuhu
Basin, along with an expansion of understorey taxa, such
as grasses and ferns (Fig.4). The full response at the Daji-
uhu Basin took many decades, and once the environmental
stress was relieved, there was a return to developed forest,
and decline in grass and fern cover, but also some changes
in forest species representation.
Due to the different pathway of CO2 fixation in the pro-
cess of photosynthesis, δ13C values have a clear distinction
between C3 and C4 plants. The δ13C values of C4 plants dis-
tribute from −20 to −9‰ with an average of −14‰. But
the δ13C values of C3 plants are more negative and vary
from −35 to −21‰ (O’Leary 1988). Thus, our results indi-
cated that the accumulative plant remnants in Dajiuhu peat
were mainly C3 plants. The δ13C values declined at around
9.1 cal ka BP, with a rapid decrease between 9.1 and
8.7cal ka BP, probably due to the shifts of vegetation and
plants composition, which matched with the pollen records
(Fig.5).
In summary, our pollen records suggested that the
peatland plants in the Dajiuhu Basin declined during
9.1–8.7cal ka BP, so did the tree cover in adjacent moun-
tains, whereas more grasses can be seen during that interval
(Figs.4, 5). The pollen and δ13C records showed that the
responses of vegetation to climate change had a time lag
of about 200–300 years. It suggested that the 9.2cal ka BP
event was at least a regional climate event in central China,
instead of an only local deposition/flood event within the
Dajiuhu Basin.
4.4 Collapse ofAsian summer monsoon precipitation
ataround9.2cal ka BP
The geochemical proxies in this study, together with the
pollen and δ13C records, provided a robust evidence for
an abrupt dry and cold climate event in the Dajiuhu Basin
and its adjacent areas around the 9.2cal ka BP, which was
probably connected to the weakening of the EASM. A
weakened EASM and decreased monsoonal precipitation
in the Dajiuhu Basin around 9.2cal ka BP were also inde-
pendently supported by adjacent stalagmite records, San-
bao (31°40′N, 110°26′E), Heshang (30°27′N, 110°25′E),
and Dongge (25°17′N, 108°5′E, Fig. 7c) Caves, whose
δ18O values expressed a dramatic peak level around 9.2cal
ka BP, indicating a significant decrease in monsoonal
precipitation (Dong et al. 2010; Dykoski et al. 2005; Hu
et al. 2008). The consistent oscillations in stalagmite
δ18O records (Dong etal. 2010; Dykoski etal. 2005; Hu
etal. 2008) and our multi-proxy peatland records around
9.2cal ka BP indicated that the 9.2ka event probably had

W.Zhang et al.
1 3
a wide-spread impact on central China with a sudden col-
lapse of monsoon precipitation.
In addition to the proxy records from central China, other
monsoon records in China also showed a weakened inten-
sity of the EASM during 9.5–9.0cal ka BP. The summer
monsoon index (SMI) from Qinghai Lake (36°48′40.7″N,
100°08′13.5″E, Fig. 7a) began declining since 9.5 cal
ka BP, with a minimum at about 9.2cal ka BP (An etal.
2012), indicating a sharp decrease in EASM intensity,
which was in excellent agreement with the reconstructed
precipitation records of Gonghai Lake (38°54′N, 112°14′E)
(Chen etal. 2015). Further, the difference in δ13C values
between C31 and C29 n-alkanes from Huguangyan Maar
Lake (21°9′N, 110°17′E, Fig.7d), as a proxy of the EASM
intensity, pointed to a dramatic weakening in EASM inten-
sity around 9.2cal ka BP (Jia etal. 2015), so did the TOC
records in Erhai Lake (25°25′–26°10′N, 99°32′–100°27′E)
(Zhou etal. 2007).
A large number of Holocene climate proxy records in
the Indian Ocean monsoon region suggest that the inten-
sity of the Indian summer monsoon (ISM) weakened rap-
idly around 9.2cal ka BP (Fleitmann et al. 2003; Gupta
et al. 2003, 2005; Neff etal. 2001). Gupta et al. (2003,
2005) used the G. bulloides proxy from the Arabian Sea
(18°03.079′N, 57°36.561′E, Fig.7e) to investigate the his-
tory of ISM changes, with results showing that the intensity
Fig. 7 The comparison
between the Rb/Sr record from
Dajiuhu peat and the other pale-
oclimate records. a The summer
monsoon index (SMI) of
Qinghai Lake (3-point running
average) (An etal. 2012); b the
Rb/Sr ratio of Dajiuhu peat in
this study; c the δ18O record of
Dongge Cave (3-point running
average) (Dykoski etal. 2005);
d the Δδ13C31−29 of Huguang-
yan Maar Lake (3-point running
average) (Jia etal. 2015); e
G. bulloides of ODP 723A
(Gupta etal. 2003, 2005); f the
Hematite-stained grains (HSG)
content of North Atlantic (Bond
etal. 2001); g the δ18O records
of NGRIP (7-point running
average) (Rasmussen etal.
2006; Vinther etal. 2006); h
the 14C production rate (7-point
running average) (Damon and
Peristykh 2000); i the 10Be flux
from GRIP and GISP2 com-
bined (7-point running average)
(Muscheler etal. 2004). All of
the records are detrended with
quadratic polynomial. The blue
bars indicate the 8.2 and 9.2ka
events

The 9.2ka event inAsian summer monsoon area: thestrongest millennial scale collapse ofthe…
1 3
of ISM became relatively weaker during 9.5–9.0 cal ka
BP. Consistently, the stalagmite δ18O records from Hoti
(23°05′N, 57°21′E) and Qunf (17°10′N, 54°18′E) Caves
illustrate a reduced amount of monsoonal precipitation
at about 9.2cal ka BP (Fleitmann et al. 2003; Neff et al.
2001).
In summary, our peat records, together with a large
body of other records from the whole ASM area, including
records from the EASM front area in northern China, the
central and southern China and the ISM area, consistently
presented a significant decrease of monsoon precipitation
around 9.2cal ka BP.
4.5 The 9.2ka event versusthe8.2ka event intheASM
area
The climate of the Holocene is marked by a series of dec-
adal-to-centennial scale abrupt climate oscillations, such
as the 9.2, 8.2, 5.0, 4.2 and 0.5 ka climate events (Fleit-
mann etal. 2008; Parker etal. 2006; Risebrobakken etal.
2003; Soon etal. 2014; Wang etal. 2005; Yu etal. 2006;
Zhang et al. 2008). These Holocene climate oscillations
have received much attention in recent decades and the
8.2ka climate event is rather well studied. The impact of
the 8.2ka event on the ASM has been widely investigated
(Fleitmann etal. 2003; Gupta etal. 2003, 2005; Liu et al.
2013; Neff etal. 2001; Raj etal. 2015). The results from
diverse proxy records suggest that both the EASM and ISM
weakened clearly and the precipitation decreased during
the 8.2ka event.
In contrast to the 8.2ka event, the impacts of the 9.2ka
event have attracted much less attention and its magnitude
and performance in the ASM area have not been fully clari-
fied. However, our multi-proxy peat records in this study
provided robust evidence for a significant abrupt climate
event in the Dajiuhu Basin and adjacent areas around
9.2cal ka BP, which was probably the largest climate oscil-
lation within the Holocene and evidently stronger than that
of the 8.2ka event (Fig.3). In addition to our peat records,
quite a few other independent hydrologic records from the
ASM area also presented a relatively larger monsoon col-
lapse around 9.2cal ka BP than that of the 8.2 ka event,
such as the lake sediment records from Gonghai, Qinghai
(Fig.7a), Erhai and Huguangyan Maar (Fig.7d) Lakes (An
etal. 2012; Chen et al. 2015; Jia etal. 2015; Zhou et al.
2007), the stalagmite records from Heshang and Dongge
Caves (Fig.7c) (Dykoski etal. 2005; Hu etal. 2008), the
marine sediment record from Arabian Sea (ODP 723A,
Fig.7e) (Gupta etal. 2003, 2005) and the peat records from
the Dajiuhu Basin (Fig.6c) (Huang 2009), all within the
margin of error in chronology. Besides, the 9.2 ka event
recorded in the Sanbao, Qunf and Huti Caves was also
slightly larger than the 8.2ka event, or at least as evident
as the 8.2ka event (Dong etal. 2010; Fleitmann etal. 2003;
Neff etal. 2001). Thus we propose that the 9.2 ka event
could be the largest climate anomaly event during the early
Holocene in the ASM area.
4.6 Possible mechanism ofthe9.2ka weak Asian
summer monsoon event
Changes in the amount of the North Atlantic fresh water
input in regions of deep water formation have been rec-
ognized as an important driver of millennial-scale abrupt
ASM weakening events during the last glacial period and
the Holocene (Cheng et al. 2012; Dykoski et al. 2008;
Gupta etal. 2003; Pausata et al. 2011; Wang etal. 2005,
2008; Yu etal. 2009). The correlations between the weak-
ened ASM around 9.2 cal ka BP (An et al. 2012; Chen
etal. 2015; Dong et al. 2010; Dykoski etal. 2005; Fleit-
mann etal. 2003; Gupta etal. 2003, 2005; Hu etal. 2008;
Jia et al. 2015; Neff et al. 2001; Zhou et al. 2007) and
the drift ice record in the North Atlantic (Fig. 7f) (Bond
etal. 2001) suggested that the North Atlantic fresh water
input around 9.2cal ka BP probably had an impact on the
ASM. The North Atlantic fresh water input could reduce
the North Atlantic thermohaline circulation (THC). Once
the THC slows down, the warm surface water will accumu-
late in the tropics and southern hemisphere. The heat bal-
ance between northern and southern hemisphere would be
destroyed and the heat budget will increase in the southern
hemisphere relative to the northern hemisphere, resulting
in a southward migration of the ITCZ, as well as a weak-
ening of the ASM (Broccoli etal. 2006; Clark etal. 2002;
Dahl etal. 2005; Ellison etal. 2006; Fleitmann etal. 2008;
Gupta etal. 2003; Wang etal. 2005; Yu etal. 2009). Such a
pattern has been supported by many model simulations and
proxy records (Broccoli etal. 2006; Chiang and Bitz 2005;
Haug etal. 2001; Strikis etal. 2012; Yu etal. 2009).
The weak ASM at about 9.2cal ka BP also coincided
with a cold period documented at high latitudes near the
Arctic Circle, such as the δ18O records from NGRIP
(Fig.7g) (Rasmussen etal. 2006; Vinther etal. 2006), chi-
ronomid assemblage records from Northeast United States,
Northwest England and Arctic Canada (Axford etal. 2009;
Hou etal. 2012; Lang etal. 2010), and the biogenic silica
record in the Alaskan Subarctic (Hu et al. 2003). Cold
Northern Hemisphere continent and Arctic around 9.2cal
ka BP could also partly contribute to the weak ASM by
weakening thermal contrast between land and ocean. The
hypothesis that increased winter snowfall weakened the
ASM of the following summer through the down-stream
over Eurasia has been noted in some model simulations and
observation-based studies (Barnett etal. 1989; Gupta etal.
2003; Meehl 1994; Overpeck etal. 1996).

W.Zhang et al.
1 3
It is clear that fresh water input in the North Atlantic
and cold climate around Arctic Circle could be the poten-
tial forcings of the ASM collapse around the 9.2cal ka BP
(Barnett etal. 1989; Broccoli etal. 2006; Chiang and Bitz
2005; Gupta etal. 2003; Meehl 1994; Yu etal. 2009). How-
ever, these forcings cannot explain the stronger collapse of
the ASM around the 9.2cal ka BP compared to the 8.2cal
ka BP. If both the 8.2 and 9.2 ka weak monsoon events
are dominated by the changes from the North Atlantic, the
more fresh water input around 8.2cal ka BP [probably ini-
tiated by flooding from an ice-dammed lake (Kleiven etal.
2008)], together with the colder Arctic temperature, prob-
ably results in stronger decline of the ASM around 8.2cal
ka BP than that around 9.2cal ka BP (Bond et al. 2001;
Kleiven etal. 2008; Rasmussen et al. 2006; Vinther et al.
2006).
In addition to North Atlantic high latitude forcing,
the change of insolation could contribute directly to the
regional monsoon variability in low latitudes (Liu et al.
2009; Yan et al. 2015). Changes in the insolation can
directly affect the thermal contrast between the continent
and ocean, thereby resulting in the variations of the mon-
soon strength (Liu et al. 2009; Yan et al. 2015). When
the effective radiation flux increases, warming over land
is much stronger than that of adjacent ocean and thus the
thermal contrast between continent and ocean is reinforced.
This increased thermal contrast further enhances the pres-
sure differences between land monsoon regions and the
surrounding oceans and therefore strengthens the monsoon
circulation and its associated rainfall (Liu etal. 2009; Yan
et al. 2015). A decrease in irradiance around 9.2 cal ka
BP, derived from the 14C production rate (Fig.7h) and the
10Be flux records in the GRIP and GISP2 ice cores (Fig.7i)
(Damon and Peristykh 2000; Muscheler etal. 2004), would
thus produce the decreased monsoon moisture transport
and thus less precipitation in the ASM area.
It is especially worth noting that the solar output
decrease around 9.2 cal ka BP was stronger than that
around 8.2 cal ka BP (Figs. 7h, i, 8) (Damon and Peri-
stykh 2000; Muscheler etal. 2004). Meanwhile, the decline
of the ASM around 9.2 cal ka BP has a slight time lag
(about 200year) relative to the decrease of the solar activ-
ity (Fig. 8), probably due to the delayed response of the
geological records to the solar activity or/and the dating
errors of the paleocliamte records. Based on the fact that
the monsoon collapse was stronger around 9.2cal ka BP
than 8.2cal ka BP and there was a stronger solar irradiance
decrease around 9.2cal ka BP (Damon and Peristykh 2000;
Fig. 8 The comparison between ASM proxy records and solar activ-
ity. Left a the summer monsoon index of Qinghai Lake (An et al.
2012); b the δ18O record of Dongge Cave (Dykoski etal. 2005); c the
Rb/Sr ratio of Dajiuhu peat in this study; d the 10Be flux from GRIP
and GISP2 combined (Muscheler etal. 2004). A ~200-year time lag
was observed between ASM records and 10Be flux. Right after sub-
tracting 200 years, the 10Be flux record (grey line) shows similar vari-
ations with the ASM proxy records (An etal. 2012; Dykoski etal.
2005; Muscheler etal. 2004). All of the records are detrended with
quadratic polynomial. The blue bars indicate the 8.2ka event and the
event between 10.5 and 9.0cal ka BP

The 9.2ka event inAsian summer monsoon area: thestrongest millennial scale collapse ofthe…
1 3
Muscheler etal. 2004), we can infer that the low-latitude
hydrological process driven by the solar activity (Liu etal.
2009; Yan etal. 2015) probably play a more important role
in the 9.2ka event. Meanwhile, the similar pattern of two
successive decreases of the solar output and monsoon col-
lapse around 10 and 9.2cal ka BP provide more evidence
for the possibility of this solar driving hypothesis (Figs.7h,
i, 8).
Thus a case can be made that the abrupt and sharp
decrease of ASM around 9.2 cal ka BP is likely to be
connected to the changes of the solar activity and Atlan-
tic Meridional Overturning Circulation (AMOC). The
decreased solar irradiance would result in weak ASM
strength by declining the monsoon moisture transport from
tropical ocean to the continent in low latitudes (Liu etal.
2009; Yan et al. 2015), while the attenuating of AMOC
induced by freshwater input in the North Atlantic would
weaken the ASM through the high-low latitude interac-
tions and southward migration of the ITCZ. However, the
meltwater pulses in the North Atlantic around 9.2ka cal BP
appear to be tied to the variations in solar irradiance (Bond
etal. 2001). Therefore, we inferred that the decline of the
solar irradiance probably lead to the abrupt decrease of
ASM around 9.2cal ka BP by driving the hydrologic cycle
in low latitudes directly (low latitude processes), as well as
resulting in the southward migration of the ITCZ indirectly
through increasing the fresh water input in the North Atlan-
tic (high-low latitude interactions).
5 Conclusions
In this study, a sediment profile from the Dajiuhu Basin in
central China was collected with several geochemical prox-
ies and a pollen analysis carried out to help improve under-
standing of the 9.2ka event in central China. The results
show that TOC and TN contents expressed an abrupt
decline around 9.2cal ka BP, with significantly increased
Al and Ti contents at the same time, indicating that the peat
development was interrupted abruptly at this time. The Rb/
Sr increased sharply at around 9.4cal ka BP and reached its
highest values at around 9.2cal ka BP, probably resulting
from the variations in the chemical weathering strength.
Meanwhile, there was a decline in tree-pollen, particularly
deciduous elements such as Quercus, and an open canopy
biome characterized with an expansion of ferns and grasses
pollen, thus suggesting a drier regional climatic condition
which was fully developed between 9.1 and 8.7cal ka BP.
All of these proxy records point to a strong drier regional
climatic event at around 9.2 cal ka BP in central China
probably linked to the rapid and significant weakening of
the EASM. In addition to our records, a large amount of
hydrological records from the ASM area suggested that a
significant decrease of monsoon precipitation at around
the 9.2cal ka BP was widespread and may be the strongest
negative Holocene anomaly in the ASM area. This abrupt
and sharp decrease in ASM around 9.2cal ka BP was likely
to be connected to the decline of the solar activity and the
AMOC.
Acknowledgements Financial support for this research was
provided by the National Natural Science Foundation of China
(NSFC) (41522305, 41403018) and the research Projects from
Chinese Academy of Sciences (QYZDB-SSW-DQC001 and
132B61KYSB20160003) and Qingdao National Laboratory for
Marine Science and Technology of China (QNLM2016ORP0202).
We wish to thank Willie Soon, Hanyang Lijiang and Jun Yang for
their help in sampling and paper polishing.
References
Alley RB, Ágústsdóttir AM (2005) The 8k event: cause and conse-
quences of a major Holocene abrupt climate change. Quat Sci
Rev 24:1123–1149. doi:10.1016/j.quascirev.2004.12.004
Alley RB, Mayewski PA, Sowers T, Stuiver M, Taylor KC, Clark
PU (1997) Holocene climatic instability: a prominent,
widespread event 8200 year ago. Geology 25:483–486.
doi:10.1130/0091-7613(1997)025<0483:HCIAPW>2.3.CO;2
An Z etal (2012) Interplay between the Westerlies and Asian mon-
soon recorded in Lake Qinghai sediments since 32ka. Sci Rep
2:619. doi:10.1038/srep00619
Anderson DE (1998) A reconstruction of Holocene climatic changes
from peat bogs in north-west Scotland. Boreas 27:208–224.
doi:10.1111/j.1502-3885.1998.tb00880.x
Andrews JT, Smith LM, Preston R, Cooper T, Jennings AE
(1997) Spatial and temporal patterns of iceberg raft-
ing (IRD) along the East Greenland margin, ca. 68°N,
over the last 14 cal ka. J Quat Sci 12:1–13. doi:10.1002/
(SICI)1099-1417(199701/02)12:1<1::AID-JQS288>3.0.CO;2-T
Axford Y, Briner JP, Miller GH, Francis DR (2009) Paleoecologi-
cal evidence for abrupt cold reversals during peak Holocene
warmth on Baffin Island, Arctic Canada. Quat Res 71:142–149.
doi:10.1016/j.yqres.2008.09.006
Barber DC etal (1999) Forcing of the cold event of 8200 years ago by
catastrophic drainage of Laurentide lakes. Nature 400:344–348.
doi:10.1038/22504
Barnett TP, Dümenil L, Schlese U, Roeckner E, Latif M
(1989) The effect of Eurasian snow cover on regional
and global climate variations. J Atmos Sci 46:661–686.
doi:10.1175/1520-0469(1989)046<0661:TEOESC>2.0.CO;2
Bond G et al (1997) A pervasive millennial-scale cycle in North
Atlantic Holocene and glacial climates. Science 278:1257–
1266. doi:10.1126/science.278.5341.1257
Bond G et al (2001) Persistent solar influence on North Atlan-
tic climate during the Holocene. Science 294:2130–2136.
doi:10.1126/science.1065680
Broccoli AJ, Dahl KA, Stouffer RJ (2006) Response of the ITCZ to
Northern Hemisphere cooling. Geophys Res Lett 33:L01702.
doi:10.1029/2005GL024546
Bügelmayer-Blaschek M, Roche DM, Renssen H, Andrews JT (2016)
Internal ice-sheet variability as source for the multi-century
and millennial-scale iceberg events during the Holocene?
A model study. Quat Sci Rev 138:119–130. doi:10.1016/j.
quascirev.2016.01.026

W.Zhang et al.
1 3
Burrows M, Heijnis H, Gadd P, Haberle S (2016) A new late Qua-
ternary palaeohydrological record from the humid tropics of
northeastern Australia. Palaeogeogr, Palaeoclimatol, Palaeoecol
451:164–182. doi:10.1016/j.palaeo.2016.03.003
Chai X (1990) Peat Geoscience (in Chinese). Geological Publishing
House, Beijing
Chambers FM, Charman DJ (2004) Holocene environmental change:
contributions from the peatland archive. Holocene 14:1–6. doi:1
0.1191/0959683604hl684ed
Chambers FM, Barber KE, Maddy D, Brew J (1997) A 5500-
year proxy-climate and vegetation record from blanket mire
at Talla Moss, Borders, Scotland. Holocene 7:391–399.
doi:10.1177/095968369700700402
Chen F, Zhu Y, Li J, Shi Q, Jin L, Wünemann B (2001) Abrupt Holo-
cene changes of the Asian monsoon at millennial-and centen-
nial-scales: Evidence from lake sediment document in Minqin
Basin, NW China. Chin Sci Bull 46:1942–1947. doi:10.1007/
BF02901902
Chen F etal (2015) East Asian summer monsoon precipitation vari-
ability since the last deglaciation. Sci Rep 5:1–11. doi:10.1038/
srep11186
Cheng H, Sinha A, Wang X, Cruz FW, Edwards RL (2012) The
Global Paleomonsoon as seen through speleothem records from
Asia and the Americas. Clim Dyn 39:1045–1062. doi:10.1007/
s00382-012-1363-7
Chiang JCH, Bitz CM (2005) Influence of high latitude ice cover on
the marine Intertropical Convergence Zone. Clim Dyn 25:477–
496. doi:10.1007/s00382-005-0040-5
Clark PU, Pisias NG, Stocker TF, Weaver AJ (2002) The role of
the thermohaline circulation in abrupt climate change. Nature
415:863–869. doi:10.1038/415863a
Dahl KA, Broccoli AJ, Stouffer RJ (2005) Assessing the role of North
Atlantic freshwater forcing in millennial scale climate vari-
ability: a tropical Atlantic perspective. Clim Dyn 24:325–346.
doi:10.1007/s00382-004-0499-5
Damon PE, Peristykh AN (2000) Radiocarbon calibration and appli-
cation to geophysics, solar physics, and astrophysics. Radiocar-
bon 42:137–150. doi:10.1017/S0033822200053108
Dixit Y, Hodell DA, Sinha R, Petrie CA (2014) Abrupt weakening
of the Indian summer monsoon at 8.2 kyr B.P. Earth Planet Sci
Lett 391:16–23. doi:10.1016/j.epsl.2014.01.026
Dong J etal (2010) A high-resolution stalagmite record of the Holo-
cene East Asian monsoon from Mt Shennongjia, central China.
Holocene 20:257–264. doi:10.1177/0959683609350393
Dykoski CA et al (2005) A high-resolution, absolute-dated Holo-
cene and deglacial Asian monsoon record from Dongge
Cave, China. Earth Planet Sci Lett 233:71–86. doi:10.1016/j.
epsl.2005.01.036
Dykoski CA, Edwards RL, Cheng H, Yuan D, Shen R (2008) Asian
monsoon millennial-scale variability during the last glacial
period and its links to North Atlantic climate. In: American
Geophysical Union, Fall Meeting
Ellison CRW, Chapman MR, Hall IR (2006) Surface and deep ocean
interactions during the cold climate event 8200 years ago. Sci-
ence 312:1929–1932. doi:10.1126/science.1127213
EPA U (2001) Method 200.7, Trace Elements in Water, Solids, and
Biosolids by Inductively Coupled Plasma-Mass Spectrometry
vol Revision 5.0, EPA-821-R-01-010. Office of Research and
Development, Cincinatti, Ohio
Ferrat M, Weiss DJ, Spiro B, Large D (2012) The inorganic geochem-
istry of a peat deposit on the eastern Qinghai-Tibetan Plateau
and insights into changing atmospheric circulation in central
Asia during the Holocene. Geochim Cosmochim Acta 91:7–31.
doi:10.1016/j.gca.2012.05.028
Fleitmann D, Burns SJ, Mudelsee M, Neff U, Kramers J, Mangini
A, Matter A (2003) Holocene forcing of the Indian monsoon
recorded in a stalagmite from southern Oman. Science
300:1737–1739. doi:10.1126/science.1083130
Fleitmann D et al (2007) Holocene ITCZ and Indian mon-
soon dynamics recorded in stalagmites from Oman and
Yemen (Socotra). Quat Sci Rev 26:170–188. doi:10.1016/j.
quascirev.2006.04.012
Fleitmann D, Mudelsee M, Burns SJ, Bradley RS, Kramers J, Matter
A (2008) Evidence for a widespread climatic anomaly at around
9.2ka before present. Paleoceanography 23:PA1102. doi:10.10
29/2007PA001519
Gibbs MT, Kump LR (1994) Global chemical erosion during the
last glacial maximum and the present: sensitivity to changes
in lithology and hydrology. Paleoceanography 9:529–543.
doi:10.1029/94PA01009
Gupta AK, Anderson DM, Overpeck JT (2003) Abrupt changes in the
Asian southwest monsoon during the Holocene and their links
to the North Atlantic Ocean. Nature 421:354–357. doi:10.1038/
nature01340
Gupta AK, Das M, Anderson DM (2005) Solar influence on the
Indian summer monsoon during the Holocene. Geophys Res
Lett 32:L17703. doi:10.1029/2005GL022685
Haug GH, Hughen KA, Sigman DM, Peterson LC, Röhl U (2001)
Southward migration of the intertropical convergence zone
through the Holocene. Science 293:1304–1308. doi:10.1126/
science.1059725
He Y, Zhao C, Zheng Z, Liu Z, Wang N, Li J, Cheddadi R (2015)
Peatland evolution and associated environmental changes in
central China over the past 40,000 years. Quat Res 84:255–261.
doi:10.1016/j.yqres.2015.06.004
Horák-Terra I, Cortizas AM, da Luz CFP, López PR, Silva AC, Vidal-
Torrado P (2015) Holocene climate change in central-eastern
Brazil reconstructed using pollen and geochemical records
of Pau de Fruta mire (Serra do Espinhaço Meridional, Minas
Gerais). Palaeogeogr Palaeoclimatol Palaeoecol 437:117–131.
doi:10.1016/j.palaeo.2015.07.027
Hou J, Huang Y, Shuman BN, Oswald WW, Foster DR (2012)
Abrupt cooling repeatedly punctuated early-Holocene cli-
mate in eastern North America. Holocene 22:525–529.
doi:10.1177/0959683611427329
Hu F et al (2003) Cyclic variation and solar forcing of Holocene
climate in the Alaskan subarctic. Science 301:1890–1893.
doi:10.1126/science.1088568
Hu C, Henderson GM, Huang J, Xie S, Sun Y, Johnson KR (2008)
Quantification of Holocene Asian monsoon rainfall from spa-
tially separated cave records. Earth Planet Sci Lett 266:221–
232. doi:10.1016/j.epsl.2007.10.015
Huang X (2009) Early diagenesis of peat lipids and their responses to
the climate changes over the past 13 ka: evidence from the Daji-
uhu peat deposit, South China. China University of Geosciences
(Wuhan)
ISO (1995) Soil quality: determination of organic and total carbon
after dry combustion (elementary analysis). ISO
Isono D, Yamamoto M, Irino T, Oba T, Murayama M, Nakamura
T, Kawahata H (2009) The 1500-year climate oscillation in
the midlatitude North Pacific during the Holocene. Geology
37:591–594. doi:10.1130/G25667A.1
Jia G etal (2015) Biogeochemical evidence of Holocene East Asian
summer and winter monsoon variability from a tropical maar
lake in southern China. Quat Sci Rev 111:51–61. doi:10.1016/j.
quascirev.2015.01.002
Jin Z, Wang S, Shen J, Zhang E, Li F, Ji J, Lu X (2001) Chemical
weathering since the Little Ice Age recorded in lake sediments:
a high-resolution proxy of past climate. Earth Surf Proc Land
26:775–782. doi:10.1002/esp.224
Jin Z, Shen J, Wang S, Zhang E (2003) Evidence for early Holocene
cold event from lake sediments. Geol J China Univ 9:11–18

The 9.2ka event inAsian summer monsoon area: thestrongest millennial scale collapse ofthe…
1 3
Jin Z, Wu J, Cao J, Wang S, Shen J, Gao N, Zou C (2004) Holocene
chemical weathering and climatic oscillations in north China:
evidence from lacustrine sediments. Boreas 33:260–266
Jin Z, Cao J, Wu J, Wang S (2006) A Rb/Sr record of catchment
weathering response to Holocene climate change in Inner Mon-
golia. Earth Surf Proc Land 31:285–291. doi:10.1002/esp.1243
Johnsen SJ et al (2001) Oxygen isotope and palaeotemperature
records from six Greenland ice-core stations: Camp Century,
Dye-3, GRIP, GISP2, Renland and NorthGRIP. J Quat Sci
16:299–307. doi:10.1002/jqs.622
Kleiven HK, Kissel C, Laj C, Ninnemann US, Richter TO, Cor-
tijo E (2008) Reduced North Atlantic deep water coeval with
the glacial Lake Agassiz freshwater outburst. Science 319:60.
doi:10.1126/science.1148924
Knaap WOVD etal (2011) A multi-proxy, high-resolution record of
peatland development and its drivers during the last millennium
from the subalpine Swiss Alps. Quat Sci Rev 30:3467–3480.
doi:10.1016/j.quascirev.2011.06.017
Koinig KA, Shotyk W, Lotter AF, Ohlendorf C, Sturm M (2003) 9000
years of geochemical evolution of lithogenic major and trace
elements in the sediment of an alpine lake-the role of climate,
vegetation, and land-use history. J Paleolimnol 30:307–320. doi
:10.1023/A:1026080712312
Korhola A, Vasko K, Toivonen HTT, Olander H (2002) Holocene
temperature changes in northern Fennoscandia reconstructed
from chironomids using Bayesian modelling. Quat Sci Rev
21:1841–1860. doi:10.1016/S0277-3791(02)00003-3
Kylander ME, Bindler R, Cortizas AM, Gallagher K, Mörth CM,
Rauch S (2013) A novel geochemical approach to paleorecords
of dust deposition and effective humidity: 8500 years of peat
accumulation at Store Mosse (the “Great Bog”), Sweden. Quat
Sci Rev 69:69–82. doi:10.1016/j.quascirev.2013.02.010
Lang B, Bedford A, Brooks SJ, Jones RT, Richardson N, Birks
HJB, Marshall JD (2010) Early-Holocene tempera-
ture variability inferred from chironomid assemblages at
Hawes Water, northwest England. Holocene 20:943–954.
doi:10.1177/0959683610366157
Li J etal (2013) Vegetation changes during the past 40,000 years in
Central China from a long fossil record. Quatern Int 310:221–
226. doi:10.1016/j.quaint.2012.01.009
Liu J, Wang B, Ding Q, Kuang X, Soon W, Zorita E (2009) Centen-
nial variations of the global monsoon precipitation in the last
millennium: results from ECHO-G model. J Climate 22:2356–
2371. doi:10.1175/2008JCLI2353.1
Liu Z et al (2014) Chinese cave records and the East Asia Sum-
mer Monsoon. Quat Sci Rev 83:115–128. doi:10.1016/j.
quascirev.2013.10.021
Liu Y, Henderson GM, Hu C, Mason AJ, Charnley N, Johnson KR,
Xie S (2013) Links between the East Asian monsoon and North
Atlantic climate during the 8200year event. Nat Geosci 6:117–
120. doi:10.1038/ngeo1708
Ma C et al (2008) High-resolution geochemistry records of cli-
mate changes since late-glacial from Dajiuhu peat in Shen-
nongjia Mountains, Central China. Chin Sci Bull 53:28–41.
doi:10.1007/s11434-008-5007-6
Ma C, Zhu C, Zheng C, Yin Q, Zhao Z (2009) Climate changes in
East China since the Late-glacial inferred from high-resolution
mountain peat humification records. Sci China Ser D 52:118–
131. doi:10.1007/s11430-009-0003-5
Mayewski PA et al (2004) Holocene climate variability. Quat Res
62:243–255. doi:10.1016/j.yqres.2004.07.001
Mcdermott F, Mattey DP, Hawkesworth C (2001) Centennial-scale
Holocene climate variability revealed by a high-resolution spe-
leothem δ18O record from SW Ireland. Science 294:1328–1331.
doi:10.1126/science.1063678
Meehl GA (1994) Influence of the land surface in the
Asian summer monsoon: external conditions ver-
sus internal feedbacks. J Climate 7:1033–1049.
doi:10.1175/1520-0442(1994)007<1033:IOTLSI>2.0.CO;2
Moore PD, Webb JA, Collison ME (1991) Pollen analysis. Blackwell
Scientific Publications
Moros M, Andrews JT, Eberl DD, Jansen E (2006) The Holocene
history of drift ice in the northern North Atlantic: Evidence
for different spatial and temporal modes. Paleoceanography
21:PA2017. doi:10.1029/2005PA001214
Muscheler R et al (2004) Changes in the carbon cycle during the
last deglaciation as indicated by the comparison of 10Be and
14C records. Earth Planet Sci Lett 219:325–340. doi:10.1016/
S0012-821X(03)00722-2
Neff U, Burns SJ, Mangini A, Mudelsee M, Fleitmann D, Matter A
(2001) Strong coherence between solar variability and the mon-
soon in Oman between 9 and 6 kyr ago. Nature 411:290–293.
doi:10.1038/35077048
Nesbitt HW, Young GM (1982) Early Proterozoic climates and plate
motions inferred from major element chemistry of lutites.
Nature 299:715–717. doi:10.1038/299715a0
O’Leary MH (1988) Carbon isotopes in photosynthesis. Bioscience
38:328–336. doi:10.2307/1310735
Overpeck J, Anderson D, Trumbore S, Prell W (1996) The southwest
Indian Monsoon over the last 18,000 years. Clim Dyn 12:213–
225. doi:10.1007/BF00211619
Owen RA, Day CC, Hu C, Liu Y, Pointing MD, Blättler CL, Hender-
son GM (2016) Calcium isotopes in caves as a proxy for aridity:
Modern calibration and application to the 8.2 kyr event. Earth
Planet Sci Lett 443:129–138. doi:10.1016/j.epsl.2016.03.027
Parker AG, Goudie AS, Stokes S, White K, Hodson MJ, Manning M,
Kennet D (2006) A record of Holocene climate change from
lake geochemical analyses in southeastern Arabia. Quat Res
66:465–476. doi:10.1016/j.yqres.2006.07.001
Pausata FSR, Battisti DS, Nisancioglu KH, Bitz CM (2011) Chinese
stalagmite δ18O controlled by changes in the Indian monsoon
during a simulated Heinrich event. Nat Geosci 4:474–480.
doi:10.1038/ngeo1169
Pross J et al (2009) Massive perturbation in terrestrial ecosystems
of the Eastern Mediterranean region associated with the 8.2
kyr B.P. climatic event. Geology 37:887–890. doi:10.1130/
g25739a.1
Raj R, Chamyal LS, Prasad V, Sharma A, Tripathi JK, Verma P
(2015) Holocene climatic fluctuations in the Gujarat Alluvial
Plains based on a multiproxy study of the Pariyaj Lake archive,
western India. Palaeogeogr Palaeoclimatol Palaeoecol 421:60–
74. doi:10.1016/j.palaeo.2015.01.004
Rasmussen SO etal (2006) A new Greenland ice core chronology for
the last glacial termination. J Geophys Res 111:D06102. doi:10
.1029/2005JD006079
Rasmussen SO, Vinther BM, Clausen HB, Andersen KK (2007)
Early Holocene climate oscillations recorded in three Green-
land ice cores. Quat Sci Rev 26:1907–1914. doi:10.1016/j.
quascirev.2007.06.015
Reimer PJ et al (2013) IntCal13 and Marine13 radiocarbon age
calibration curves 0–50,000 years cal BP. Radiocarbon
55:1869–1887
Risebrobakken B, Jansen E, Andersson C, Mjelde E, Hevrøy K
(2003) A high-resolution study of Holocene paleoclimatic and
paleoceanographic changes in the Nordic Seas. Paleoceanogra-
phy 18:123–126. doi:10.1029/2002PA000764
Schulz M, Paul A (2002) Holocene climate variability on centennial-
to-millennial time scales: 1. Climate records from the North-
Atlantic realm. In: Climate development and history of the
North Atlantic realm. Springer, pp41–54

W.Zhang et al.
1 3
Shao X, Wang Y, Cheng H, Kong X, Wu J, Edwards RL (2006)
Long-term trend and abrupt events of the Holocene Asian mon-
soon inferred from a stalagmite δ18O record from Shennon-
gjia in Central China. Chin Sci Bull 51:221–228. doi:10.1007/
s11434-005-0882-6
Soon W etal (2014) A review of Holocene solar-linked climatic vari-
ation on centennial to millennial timescales: Physical processes,
interpretative frameworks and a new multiple cross-wavelet
transform algorithm. Earth Sci Rev 134:1–15. doi:10.1016/j.
earscirev.2014.03.003
Spurk M, Leuschner HH, Baillie MGL, Briffa KR, Friedrich M
(2002) Depositional frequency of German subfossil oaks: Cli-
matically and non-climatically induced fluctuations in the Holo-
cene. Holocene 12:707–715. doi:10.1191/0959683602hl583rp
Stager JC, Mayewski PA (1997) Abrupt early to mid-Holocene cli-
matic transition registered at the equator and the poles. Science
276:1834–1836. doi:10.1126/science.276.5320.1834
Strikis NM etal (2012) Abrupt variations in South American mon-
soon rainfall during the Holocene based on a speleothem
record from central-eastern Brazil. Geology 39:1075–1078.
doi:10.1130/G32098.1
Vinther BM etal (2006) A synchronized dating of three Greenland
ice cores throughout the Holocene. J Geophys Res 111:D13102.
doi:10.1029/2005jd006921
Wang N, Yao T, Thompson L, Henderson K, Davis M (2002) Evi-
dence for cold events in the early Holocene from the Guliya ice
core, Tibetan Plateau, China. Chin Sci Bull 47:1422–1427
Wang Y etal (2005) The Holocene Asian monsoon: Links to solar
changes and North Atlantic climate. Science 308:854–857.
doi:10.1126/science.1106296
Wang Y etal (2008) Millennial-and orbital-scale changes in the East
Asian monsoon over the past 224,000 years. Nature 451:1090–
1093. doi:10.1038/nature06692
Wanner H, Solomina O, Grosjean M, Ritz SP, Jetel M (2011) Struc-
ture and origin of Holocene cold events. Quat Sci Rev 30:3109–
3123. doi:10.1016/j.quascirev.2011.07.010
Wu J, Wang S, Wang H (1995) Characters of the evolution of climate
and environment of Holocene in Aibi Lake basin in Xinjiang.
Oceanol Limnol Sin 27:524–536
Xue J, Zhong W, Xie L, Unkel I (2015) Vegetation responses to the
last glacial and early Holocene environmental changes in the
northern Leizhou Peninsula, south China. Quat Res 84:223–
231. doi:10.1016/j.yqres.2015.08.001
Yan H etal (2015) Dynamics of the intertropical convergence zone
over the western Pacific during the Little Ice Age. Nat Geosci
8:315–320. doi:10.1038/ngeo2375
Yu Y etal (2006) Millennial-scale Holocene climate variability in
the NW China drylands and links to the tropical Pacific and
the North Atlantic. Palaeogeogr Palaeoclimatol Palaeoecol
233:149–162. doi:10.1016/j.palaeo.2005.09.008
Yu L, Gao Y, Wang H, Guo D, Li S (2009) The responses of East
Asian Summer monsoon to the North Atlantic Meridional Over-
turning Circulation in an enhanced freshwater input simulation.
Chin Sci Bull 54:4724–4732. doi:10.1007/s11434-009-0720-3
Zhang P etal (2008) A test of climate, sun, and culture relationships
from an 1810-year Chinese cave record. Science 322:940–942.
doi:10.1126/science.1163965
Zhang W etal (2016) Peatland development and climate changes in
the Dajiuhu basin, central China, over the last 14,100 years.
Quatern Int 425:273–281. doi:10.1016/j.quaint.2016.06.039
Zhou J, Wang S, Yang G, Xiao H (2007) Younger Dryas event and
cold events in early-mid Holocene: record from the sediment of
Erhai Lake. Adv Clim Change Res 3:41–44