Page 1
Simulation of the East Asian Summer Monsoon during the Last Millenniumwith the MPI Earth System Model
WENMIN MAN
LASG, Institute of Atmospheric Physics, Chinese Academy of Sciences, and Graduate
University of Chinese Academy of Sciences, Beijing, China
TIANJUN ZHOU
LASG, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
JOHANN H. JUNGCLAUS
Max Planck Institute for Meteorology, Hamburg, Germany
(Manuscript received 20 August 2011, in final form 16 May 2012)
ABSTRACT
The decadal–centennial variations of East Asian summer monsoon (EASM) and the associated rainfall
change during the past millennium are simulated using the earth system model developed at the Max Planck
Institute for Meteorology. The model was driven by up-to-date reconstructions of external forcing including
the recent low-amplitude estimates of solar variations. Analysis of the simulations indicates that the EASM is
generally strong during the Medieval Warm Period (MWP; A.D. 1000–1100) and weak during the Little Ice
Age (LIA; A.D. 1600–1700). The monsoon rainband exhibits a meridional tripolar pattern during both ep-
ochs. Excessive (deficient) precipitation is found over northern China (358–428N, 1008–1208E) but deficient(excessive) precipitation is seen along the Yangtze River valley (278–348N, 1008–1208E) during the MWP
(LIA). Both similarities and disparities of the rainfall pattern between the model results herein and the proxy
data have been compared, and reconstructions from Chinese historical documents and some geological ev-
idence support the results. The changes of the EASM circulation including the subtropical westerly jet stream
in the upper troposphere and the western Pacific subtropical high (WPSH) in the middle and lower tropo-
sphere are consistent with the meridional shift of the monsoon rain belt during both epochs. The meridional
monsoon circulation changes are accompanied with anomalous southerly (northerly) winds between 208 and508N during the MWP (LIA). The land–sea thermal contrast change caused by the effective radiative forcing
leads to the MWP and LIA monsoon changes. The ‘‘warmer land–colder ocean’’ anomaly pattern during the
MWP favors a stronger monsoon, while the ‘‘colder land–warmer ocean’’ anomaly pattern during the LIA
favors a weaker monsoon.
1. Introduction
The East Asian summer monsoon (EASM) is an im-
portant component in the global climate system. Its
anomalous behavior leads to deficient or excessive pre-
cipitation and hence causes great economic and social
losses in East Asian regions [see reviews by Wang (2006)
andZhou et al. (2009a)]. TheEASMexhibits considerable
variability on a wide range of time scales. Many studies
have focused on the interannual or interdecadal variability
of the EASM (Chang et al. 2000a,b; Wang et al. 2000;Wu
et al. 2003; Yu et al. 2004; Wu et al. 2009a,b; Zhou et al.
2009b; Li et al. 2010). However, the behavior of mon-
soon variability on the decadal–centennial time scale
during the last millennium is less examined and largely
unknown.
Proxy data derived from Chinese historical documents
and speleothem records have been used to reconstruct
the past EASM variability, as well as the spatial patterns
and temporal evolutions of precipitation over eastern
China (Wang et al. 1987; Qian et al. 2003; Zheng et al.
Corresponding author address: Dr. Tianjun Zhou, LASG, In-
stitute of Atmospheric Physics, Chinese Academy of Sciences,
Beijing 100029, China.
E-mail: [email protected]
7852 JOURNAL OF CL IMATE VOLUME 25
DOI: 10.1175/JCLI-D-11-00462.1
� 2012 American Meteorological Society
Page 2
2006; Zhang et al. 2008). A dataset from a 120-station
drought/flood (D/F) index (a five-grade category index)
was derived fromChinese historical documents for A.D.
1470–1979 (CMA 1981). Based on the documentary
data, Wang et al. (1981) reported that the anomaly of
summer rainfall over northern China (358–428N, 1008–1208E) was usually opposite to that over the lower-
middle Yangtze River valley (278–348N, 1008–1208E).Based on the extended dataset of the D/F index for the
period of A.D. 950–1991, it was inferred that there was
more flooding in northern China during the Medieval
Warm Period (MWP), while a similar condition was
found along the Yangtze River valley during the Little
Ice Age (LIA) (Wang et al. 1987). The flood frequency
anomalies over theYellowRiver valley (338–408N, 1058–1208E) and southern China (228–298N, 1088–1208E) in
the warm season (May–September) during A.D. 991–
1999 indicate that the flood frequency was small (1.15
decade21) over the Yellow River valley and large (2.45
decade21) over southern China during A.D. 1400–1600,
which corresponds to a weakened EASM in this period.
The flood frequency markedly increased over the Yel-
low River valley and decreased over southern China
after A.D. 1650, indicating a stronger EASM (Qian et al.
2003). Zheng et al. (2006) found that the precipitation
variation in eastern China exhibited dry/wet fluctuations
on centennial time scales. Droughts dominated in the
twelfth to fourteenth centuries, but since the middle of
the seventeenth century eastern China has been more
subject to flooding. However, Zhang et al. (2008) in-
dicated that the EASM was strong during the MWP and
generally wet over eastern China, whereas the LIA was
characterized by a period of weak EASM and drought
over all of eastern China. The Swiss record of Alpine
glaciation also captures a generally strong summer Asian
monsoon during the MWP and the prominent glacial
advances correlate with a weakening summer monsoon
during the LIA (Holzhauser et al. 2005). Since the recon-
structions have to rely on relatively sparse data sources,
there exist controversial issues in understanding the
EASM variability and the associated rainfall patterns
over eastern China.
Climate models driven by external forcing agents can
provide important insights for detecting the monsoon
variability on the decadal–centennial time scale. Studies
of these issues will improve our understanding of the
physical processes that determine the long-term mon-
soon variations. These kinds of simulations have been
done using a wide range of climate models with different
levels of complexity for the last millennium (Crowley
2000; Gerber et al. 2003; Gonzalez-Rouco et al. 2003;
Goosse et al. 2005; Ammann et al. 2007; Peng et al. 2009;
Jungclaus et al. 2010; Servonnat et al. 2010). However,
previous studies of model–data intercomparison mainly
focused on variations of surface temperatures (Stouffer
et al. 2000; Zorita et al. 2005; Wagner et al. 2005; Zhang
et al. 2011) andmajor modes of climate variation such as
ENSO (Mann et al. 2005) or the North Atlantic Oscil-
lation (Shindell et al. 2003). There has been little attempt
to interpret the causes and dynamics of the decadal–
centennial EASMvariations using themodel simulations.
Liu et al. (2011) investigated the centennial–millennial
variation of the EASM precipitation over the past 1000
years through the analysis of a millennium simulation of
the coupled ECHAM and the global Hamburg Ocean
Primitive Equation (ECHO-G) model. The model re-
sults indicate that the centennial–millennial variation of
the EASM is essentially a forced response to the external
radiative forcing. The climate response of EASM to the
external radiative forcing depends on latitude. However,
multimodel intercomparisons are still needed to inves-
tigatewhether this phenomenon ismodel dependent. The
present study aims to examine the EASM changes during
the MWP and LIA by using the millennium simulations
with a comprehensive earth system model that, in con-
trast to previous studies, offers an (albeit small) ensemble
of simulations over the last millennium. The main moti-
vation of the study is to address the following questions:
1)What are the spatial structures of theEASMduring the
MWP and LIA? 2) What are the forced responses of the
East Asian summer rainfall during the two epochs? How
about the consistency between the simulation and proxy
data? 3) What is the dominant reason for the centennial
EASM changes? The remainder of the paper is organized
as follows. Section 2 provides a description of the model
and the experimental design, as well as the details of ex-
ternal forcings used in the simulations. Section 3 presents
the results. The conclusions are given in Section 4 along
with a discussion.
2. Model and data description
a. Model description
The present study is based on the millennium exper-
iments using the Max Planck Institute for Meteorology
Earth SystemModel (MPI-ESM) (Jungclaus et al. 2010).
The model consists of the atmospheric general circula-
tionmodel ECHAM5 (Roeckner et al. 2003) and theMax
Planck Institute OceanModel (MPI-OM;Marsland et al.
2003). Modules for land vegetation [Jena Scheme for
Biosphere–Atmosphere Coupling in Hamburg (JSBACH);
Raddatz et al. 2007) and ocean biogeochemistry [Hamburg
Model of the Ocean Carbon Cycle (HAMOCC);Wetzel
et al. 2006] enable the interactive simulation of the car-
bon cycle. ECHAM5 is run at T31 resolution (;3.758)
15 NOVEMBER 2012 MAN ET AL . 7853
Page 3
with 19 vertical levels, resolving the atmosphere up to
10 hPa. MPI-OM applies a conformal mapping grid
with a horizontal resolution ranging from 22 to 350 km.
The ocean model includes a Hibler-type dynamic–
thermodynamic sea icemodel with viscous plastic rheology
(Hibler 1979). Ocean and atmosphere are coupled daily
without flux corrections using the Ocean Atmosphere Sea
Ice Soil version 3 (OASIS3) coupler (Valcke et al. 2003).
b. Experimental design and forcing data
The experimental strategy is described as follows.
After a multicentury spinup phase in which the carbon
cycle was brought into equilibrium, a 3000-yr unforced
control experiment was performed under A.D. 800 or-
bital conditions and preindustrial greenhouse gas con-
centrations. Starting fromdifferent ocean initial conditions,
a five-member ensemble (E1) with the standard external
forcing spanning A.D. 800–2005 was conducted using
the earth system model.
The total solar irradiance (TSI) forcing used as the stan-
dard forcing exhibits an increase of 0.1% (;1.3 W m22)
from the Maunder Minimum to today (Viera et al. 2011),
which is in agreement with other recent evaluations, al-
though other reconstructions with higher long-term varia-
tions also exist [see the discussion in Schmidt et al. (2011)].
The volcanic forcing is calculated online in the model
using time series of aerosol optical depth and of the ef-
fective radius (Crowley et al. 2008). Anthropogenic land
cover change is considered by applying the reconstruction
of global agricultural areas and land cover (Pongratz et al.
2008). While the CO2 concentration is calculated in-
teractively within the model, the concentrations of the
other two major greenhouse gases, methane (CH4) and
nitrous oxide (N2O), are prescribed (MacFarling Meure
et al. 2006). Some other potentially important forcings
such as the orbital forcing and anthropogenic tropo-
spheric sulfate aerosols are also included in the ensemble
experiments. The orbit forcing has little effect on the
magnitude of the seasonal cycle [see Jungclaus et al.
(2010) for details].
Effective radiative forcings (Fig. 1) are calculated
offline with the ECHAM5 isolated radiative transfer
code following the Wetherald and Manabe (1998) ap-
proach for calculating radiative feedbacks. The anom-
alous total radiative forcing, which represents the sum of
the solar forcing and the radiative effects of volcanic
aerosols, land cover change, and the CO2 concentration,
follows a high value during the MWP and a low value
during the LIA in the simulations (Fig. 1).
c. Data
The data used for validation of the model performance
under the present-day climate include the following:
1) The precipitation dataset compiled by Climate Pre-
diction Center (CPC) Merged Analysis of Precipita-
tion (CMAP) for the period of 1979–2005 on a 2.58 32.58 grid (Xie and Arkin 1995); and
2) The National Centers for Environmental Prediction
(NCEP) reanalysis 2 data (NCEP2) for 1979–2005 on
a 2.58 3 2.58 grid (Kanamitsu et al. 2002).
3. Results
We first present the summer [June–August (JJA)]
rainfall distribution and seasonal cycle of precipitation
on an observational basis for the model validation. Then
we focus on the large-scale monsoonal circulation and
precipitation changes during the MWP and LIA. Fi-
nally, we try to attribute the underlying causes of the
EASM variations from the perspective of land–sea
thermal contrast. In the following discussions, all the
anomalies are calculated relative to the climate mean of
the whole millennium.
FIG. 1. Effective radiative forcing anomalies at the top of the
atmosphere (W m22) associated with variations in solar irradiance
(black), volcanic aerosols (blue), CO2 emissions (orange), and land
cover changes (green). Anomalies from solar irradiance and CO2
variations are calculated relative to their preindustrial control
mean (1367 W m22 and 280 ppm, respectively). Solar forcing has
taken into account of Earth’s cross-sectional surface area (divided
by 4) and the contribution from the planetary albedo (simply use
a constant of 0.7).
7854 JOURNAL OF CL IMATE VOLUME 25
Page 4
a. Validation of the model performance
Faithful analysis of the EASM response to external
forcing should be based on rigorous verification of the
model performance (Liu et al. 2011). Comparisons
have demonstrated that the simulated Northern Hemi-
sphere temperature evolution and regional temperature
anomalies in China agree well with the reconstructions
(Jungclaus et al. 2010; Zhang et al. 2011).We focus on the
mean state of summer rainfall as well as the seasonal
march of the monsoon rain belt to evaluate the perfor-
mance of the model in monsoon simulation in this study.
Variations of the EASM are usually described using sum-
mer rainfall and low-level (850 hPa) winds (Zhou and
Yu 2005; Yu and Zhou 2007; Chen et al. 2010). Figures 2a
and 2b compare the simulated summer rainfall and 850-
hPa winds for the average of 1979–2005 with the observa-
tions. The observed precipitation data are derived from
CMAP, while the circulation data are from NCEP2. The
summer 850-hPa winds feature strong southwesterlies
from the Indian monsoon and southeasterlies from the
western Pacific, as well as the cross-equator flow around
1058–1208E (Fig. 2a). The observed summer precipitation
generally decreases northwestward from the Southeast
Asian marginal seas toward the arid central continental
Asia, and there is a monsoon rainband extending from
the East Asian marginal continents to Japan (Fig. 2a).
The model captures the main features of the observed
summer precipitation realistically except that it under-
estimates themonsoon rainband extending from eastern
China to Japan (Fig. 2b). The monsoon wind penetrates
into northern China in the model, resulting in a north-
ward bias of the monsoon rainband. An artificial rainfall
center located to the eastern periphery of the Tibetan
Plateau is also evident in the simulation, as in many at-
mospheric general circulation models (AGCMs) (Yu
et al. 2000; Zhou and Li 2002; Chen et al. 2010). The
pattern correlation coefficient of the summer precip-
itation between the observation and model simulation is
0.81, and the root-mean-square difference between the
observation and the simulation is 2.24 mm day21.
East Asian precipitation exhibits a robust seasonal
cycle associated with monsoon development (Zhou et al.
2009a). The seasonal cycles of extratropical (368–508N,
FIG. 2. (a),(b) Mean state of JJA precipitation (colored shading; mm day21) and 850-hPa winds (vectors; m s21)
and (c),(d) seasonal cycle of precipitation (mm day21). (a) CMAP (precipitation) and NCEP2 (850-hPa winds) data,
(b)model simulation, (c) seasonal cycle of regional (368–508N, 1008–1208E) average of precipitation, and (d) seasonalcycle of regional (218–358N, 1008–1208E) average of precipitation. All figures are for the average of 1979–2005.
15 NOVEMBER 2012 MAN ET AL . 7855
Page 5
1008–1208E) and subtropical (218–358N, 1008–1208E)precipitation are shown in Figs. 2c and 2d. In extratropical
East Asia (EA; Fig. 2c), the seasonal cycle is well simu-
lated with a peak rainy month in July and a minimum in
January, but the simulated spring rainfall is stronger than
that in the observation. In subtropical East Asia (Fig. 2d),
the simulation resembles the observation in the June peak
but the strength of precipitation prior to June is over-
estimated. Overall, the model performs well in the sim-
ulation of both the seasonal cycle and precipitation
amounts in different latitudinal regions of East Asia.
b. Response of EASM circulation and precipitationduring the MWP and LIA
Since the focus of the study is the dynamic structure and
physical processes of EASM variations on the decadal–
centennial time scale, we examine the features during the
MWP (A.D. 1000–1100) and LIA (A.D. 1600–1700),
which represent two typical climatic epochs of the last
millennium in China.We focus on the ensemblemean of
five realizations in the following analysis.
Anomalies of JJA mean 850-hPa winds and pre-
cipitation during the MWP and LIA are shown in Fig. 3.
The MWP (LIA) is characterized by the strengthening
(weakening) of the 850-hPa southwesterly winds, in-
dicating a generally stronger (weaker) EASM. This re-
sult is consistent with the reconstruction from a stalagmite
record in the Wanxiang Cave, China, which indicates
a strong (weak) EASM during the MWP (LIA) (Zhang
et al. 2008).
The anomalies of monsoon rainfall exhibit a meridio-
nal tripolar pattern during both epochs. Excessive (de-
ficient) precipitation is evident over northern China
(358–428N, 1008–1208E) whereas deficient (excessive)
precipitation is seen along the Yangtze River valley
(278–348N, 1008–1208E) during the MWP (LIA). The
rainfall patterns are consistent with the reconstructions
from Chinese historical documents (Wang et al. 1987;
Qian et al. 2003), which suggest that changes of rainfall
along the Yangtze River valley are generally out of phase
from those over northern China, but different from the
reconstruction of the Wanxiang record (Zhang et al.
2008), with wet (dry) conditions over all of eastern China
during the MWP (LIA). The reason for this discrepancy
deserves further investigations. Some geological evidence
also supports more precipitation over northern China
during theMWP (Ren and Zhang 1996;Wu and Lu 2005;
Cao et al. 2004). For the rainfall along the Yangtze River
valley, the reconstructed 3000-yr precipitation curve from
the Longgan Lake (298509–308059N, 1158559–1168209E)shows that a drier climate locally occurred in most of the
MWP (Tong et al. 1997). Our result is also in accordance
with the modern definitions of the EASM in that when
southerly monsoon penetrates deeply into northern
China, extratropical EA has plentiful rainfall, but the
subtropical rainfall (known as mei-yu in China, Baiyu in
Japan, and Changma in Korea) tends to be suppressed
(Ding 1992). Note this definition is different to that
from the stalagmite proxy data (i.e., a strong EASM
often means an abundant mei-yu; Wang et al. 2005).
Our model result agrees with the observed structure on
interannual time scale. We also calculate the ensemble
spread of precipitation during the two epochs, and define
the signal-to-noise ratio as the absolute value of the en-
semble mean value dividing the ensemble spread (Zhou
and Yu 2006). The signal-to-noise ratio is greater than
1.0 over most parts of the EA region (not shown), in-
dicating that the magnitude of the forced signal is larger
than the model spread, which support the robustness of
the responses in our simulation.
FIG. 3. Anomalies of JJA mean precipitation (colored shading;
mm day21) and 850-hPa winds (vectors; m s21) for (a) the MWP
and (b) the LIA. The anomalies are calculated relative to the
millennial mean value. The regions shaded by black dots denote
areas with precipitation that are statistically significant at the 5%
level by using a Student’s t test.
7856 JOURNAL OF CL IMATE VOLUME 25
Page 6
c. The horizontal circulation of EASM during theMWP and LIA
The water vapor transport is crucial to monsoon
rainfall. Since the vertically integrated water vapor
transport is dominated by the lower troposphere (Zhou
and Yu 2005), an analysis of water vapor transport at
850 hPa is reasonable. There are three main branches
of climatological 850-hPa water vapor transport to EA
(Fig. 4a): a strong transport by the southwesterlies from
the Indian monsoon, a moderate transport by the South-
east Asian monsoon from the western Pacific, and a weak
transport of cross-equator flow straddling 1058–1508E.Anomalies of JJA mean 850-hPa water vapor transport
during the MWP and LIA are shown in Figs. 4b and 4c.
The anomalous water vapor transport by both the south-
westerlies from the Indian monsoon and the southeast-
erlies from the western Pacific are positive during the
MWP and, in addition to the anomalous southerly winds
over EA, the northward transport of tropical water vapor
to northern China has enhanced, but the water vapor
convergence over the Yangtze River valley has decreased
(Fig. 4b). This leads to excessive rainfall in northernChina
and deficient rainfall in central China along the Yangtze
River valley. The northward moisture transport has re-
duced during the LIA and there is more water vapor
convergence over the Yangtze River valley (Fig. 4c), re-
sulting in excessive rainfall in central China but deficient
rainfall in northern China.
To quantify the water vapor transports, estimations of
the atmospheric water budget across the four bound-
aries of northern China and the Yangtze River valley
have been calculated according to Li et al. (2009). During
the MWP, the 850-hPa net influx in northern China was
6.25 3 106 kg s21, which was much larger than that over
the Yangtze River valley (2.123 106 kg s21), resulting in
excessive rainfall in northern China but deficient rainfall
in central China along the Yangtze River valley. During
the LIA, the 850-hPa net influx in northern China (1.213106 kg s21) was much smaller than that over the Yangtze
River valley (6.05 3 106 kg s21). This leads to deficient
rainfall in northern China and excessive rainfall along the
Yangtze River valley.
The water vapor transport is closely linked to the mon-
soon circulation change. Previous studies have found that
the EASM is greatly controlled by the western Pacific
subtropical high (WPSH) in the middle and lower tro-
posphere and the subtropical westerly jet stream in the
upper troposphere (Tao and Chen 1987). The position,
shape, and strength of the WPSH dominate the large-
scale quasi-stationary frontal and associated rainband in
EA (Tao and Chen 1987; Ding 1994; Zhou and Yu 2005).
TheWPSH is conventionallymeasured by the geopotential
height at 500 hPa (Tao and Chen 1987; Ding 1994; Zhou
et al. 2009c). In Fig. 5, we present the changes of geo-
potential height of 500 hPa for both the MWP and LIA.
It is clear that the ridge of the geopotential isolines shift
northward (southward) during the MWP (LIA) com-
pared with the climate mean position (Figs. 5a,b). The
northward (southward) shift of the WPSH ridge during
the MWP (LIA) leads to stronger (weaker) southerlies
penetrating into northern China, resulting in deficient
(excessive) precipitation along the Yangtze River Val-
ley but excessive (deficient) precipitation over northern
China.
At the upper levels (200 hPa), the subtropical west-
erly jet is an important part of the Tibetan high, the
position and strength of which is closely related to the
EASM rainfall (Zhang et al. 2006). The most out-
standing features of the climatological EASM at 200 hPa
FIG. 4. JJAmean 850-hPawater vapor transport (kg m21 s21) for
(a) the millennial mean, (b) the MWP, and (c) the LIA. The
anomalies of the MWP and LIA are calculated relative to the
millennial mean value. The gray shading denotes regions with
meridional components that are statistically significant at the 5%
level by using a Student’s t test.
15 NOVEMBER 2012 MAN ET AL . 7857
Page 7
are the westerly jet stream centered along 408N and the
tropical easterly jet to the south of 258N (Fig. 6a). Anom-
alies of JJA mean 200-hPa zonal winds during the MWP
and LIA are shown in Figs. 6b and 6c. The subtropical
westerly jet exhibits apparent meridional shifts during the
two epochs. There is a significantly weakened (inten-
sified) westerly south to the jet axis and an intensified
(weakened) westerly north to the jet axis during the
MWP (LIA). This corresponds to the northward (south-
ward) shift of the monsoon rain belt and is generally ac-
companied by excessive (deficient) rainfall over northern
China but deficient (excessive) rainfall along the Yangtze
River valley (Lau et al. 1988; Liang and Wang 1998; Li
et al. 2004; Zhou and Yu 2005).
d. Meridional monsoon circulation changes duringthe MWP and LIA
The Hadley cell is a thermally driven meridional cir-
culation, which is characterized by a rising motion in the
tropics and a descending motion in the subtropics. The
normal Hadley cell in the East Asian monsoon region is
replaced by a meridional circulation of the opposite
sense, which is often referred to as the monsoonal me-
ridional cell (Chen et al. 1964; Ye and Yang 1979) and
has been used as observational metric in model evalu-
ations (Zhou and Li 2002; Chen et al. 2010). To examine
the meridional structure of the EASM during the MWP
and LIA, Fig. 7 shows a meridional–vertical cross sec-
tion of JJA mean winds along 1058–1228E. The clima-
tological map exhibits a strong upward motion over
Southeast Asia (;58–308N) and a weak ascent north of
about 408N(Fig. 7a). The anomalousmeridionalmonsoon
FIG. 5. Spatial distributions of JJA mean geopotential height at
500 hPa for (a) the MWP and (b) the LIA (both shown in color).
The results of the climatological millennial mean are shown in
long-dashed black. Contour interval is 10 gpm.
FIG. 6. JJA mean 200-hPa zonal wind (m s21) for (a) the mil-
lennial mean, (b) the MWP, and (c) the LIA. The anomalies of the
MWPandLIA are calculated relative to themillennial mean value.
The shading denotes regions that are statistically significant at the
5% level by using a Student’s t test.
7858 JOURNAL OF CL IMATE VOLUME 25
Page 8
circulation during the MWP shows upward motion over
the extratropical EA and downward motion over the
subtropical EA. This corresponds to anomalous low-level
southerlies between 208 and 508N (Fig. 7b). Anomalous
descending and ascending motion respectively dominate
the northern land and southern ocean areas during the
LIA, with anomalous northerly winds between 208 and508N (Fig. 7c). The meridional circulation changes in the
monsoon region are consistent with an enhanced (weak-
ened) EASM during the MWP (LIA).
e. The driving mechanisms of EASM change:Land–sea thermal contrast
The occurrence of the EASM variability is a conse-
quence of the atmospheric response to the diabatic
heating between the ocean and the land (Li and Yanai
1996). The spatial structure of the monsoon response can
be understood in terms of the effects of land–sea thermal
contrast. Studies of the monsoon variability from this
perspective can help us understand the physical and dy-
namical processes that determine the long-termmonsoon
variations.
The land–sea thermal contrast is attributable to ex-
ternal forcings. We term the effective radiative forcing
from the external drivers as the sum of the effective solar
irradiance, the radiative effects of volcanic aerosols, the
land-cover changes, and greenhouse-gas forcing. All of
the individual effective radiative forcings are calcu-
lated offline with the ECHAM5 isolated radiative
transfer code. The effective solar irradiance has been
multiplied by planetary albedo (we simply use a con-
stant of 0.7) and divided by 4. The effective radiative
forcing anomaly between the MWP and LIA differ by
0.23 W m22 in the simulation. The calculation of the in-
dividual component differs by 0.03 W m22 for the solar
forcing and 0.19 W m22 for the volcanic forcing over the
100-yr periods for the MWP and LIA. The result in-
dicates that volcanic forcing has a major contribution for
the total effective radiative forcing difference between
the MWP and LIA. The anomaly of the individual ef-
fective radiative forcing is20.01 W m22 for solar forcing
and 20.26 W m22 for volcanic forcing over the 100-yr
periods for the LIA compared with the long-term mean.
The calculation of the mean radiative forcing for the
FIG. 7. Latitude–height cross section of JJA mean meridional
circulation averaged over 1058–1228E (units of the vertical and
meridional velocity are 21024 hPa s21 and m s21, respectively)
for (a) the millennial mean, (b) the MWP, and (c) the LIA. The
anomalies of the MWP and LIA are calculated relative to the
millennial mean value. The gray shading denotes regions with
vertical components that are statistically significant at the 5% level
by using a Student’s t test.
15 NOVEMBER 2012 MAN ET AL . 7859
Page 9
LIA further indicates that the LIA is dominated by the
volcanic forcing. When the effective radiative forcing
increases during the MWP because of the different
thermal capacity of the land and ocean, the temperature
increases over theEastAsian continent, especially in the
midlatitude, more rapidly than over the adjacent ocean,
which produces the land–sea thermal contrast during the
period.
The tropospheric mean temperature is a reasonable
indicator of thermal contrast change (Zhou and Zou
2010). The tropospheric mean (200–500-hPa average)
temperature anomalies during the MWP and LIA are
shown in Fig. 8. Warm temperature anomalies prevail
over EA with a central magnitude of 0.28C during the
MWP, while cool anomalies are seen in the tropical
western Pacific and extratropical North Pacific with a
central value of 20.258C (Fig. 8a). There exist cold
anomalies with an amplitude up to20.38C over EA and
warming anomalies with a central magnitude of 0.258Cin the tropical western Pacific and extratropical North
Pacific during the LIA (Fig. 8b). The magnitude of tem-
perature changes during the LIA is stronger than the
MWP. The position of the cooling center over EA ex-
hibits a northward displacement during the LIA. Since
the mean state of summertime tropospheric mean tem-
perature features a ‘‘warm land–cold ocean’’ condition,
the ‘‘warmer land–colder ocean’’ anomaly pattern during
the MWP favors a strong EASM circulation, while the
‘‘colder land–warmer ocean’’ anomaly pattern during the
LIA favors a weak EASM circulation. Thus the land–sea
thermal contrast change is the fundamental driver of
EASM changes.
The EASM change is caused by both zonal thermal
contrast between EA and the North Pacific and merid-
ional thermal contrast between EA and the tropical
western Pacific (Fig. 8). To have a clear picture of both
zonal and meridional land–sea thermal contrast change,
the corresponding structures at vertical cross sections are
shown in Figs. 9 and 10. The zonal land–sea thermal
contrast along 308–458Ndepicted by the height–longitude
cross section (Fig. 9) exhibits a signature of warmer land–
colder ocean during theMWP, with warmer anomalies of
0.28C over the Eurasian continent extending from 608 to1208E. The cooler anomalies with a central magnitude of
20.28C are seen over the ocean area in the middle-upper
troposphere, extending from 1208E to 1508W. Tempera-
ture anomalies of almost reversed sign are evident during
the LIA, with negative anomalies over land extending
from 758 to 1208E and positive anomalies with an am-
plitude up to 0.38C over an ocean area extending from
FIG. 8. JJA mean upper-tropospheric (500–200 hPa) tempera-
ture anomalies (8C) for (a) the MWP and (b) the LIA. The
anomalies are calculated relative to the millennial mean value. The
black stippled regions denote areas that are statistically significant
at the 5% level by using a Student’s t test.FIG. 9. Longitude–height cross section of JJA temperature av-
eraged over 308–458N (8C) for (a) the MWP and (b) the LIA. The
anomalies are calculated relative to the millennial mean value. The
black stippled regions denote areas that are statistically significant
at the 5% level by using a Student’s t test.
7860 JOURNAL OF CL IMATE VOLUME 25
Page 10
1208E to 1508W. The simulated colder land is weaker in
magnitude (;0.18C) and narrower in zonal extent, ex-
hibiting the maximum magnitude in the middle-lower
troposphere below 500 hPa. However, the warmer anom-
alies over ocean area show a deep vertical structure and
penetrate throughout the troposphere.
The height–latitude cross section measuring the me-
ridional land–sea thermal contrast also exhibits a warmer
land–colder ocean structure during the MWP (Fig. 10a).
The ‘‘warmer land’’ is evident with its maximumwarming
of 0.38C around 300–500 hPa extending from 308 to 608N.
The temperature over the ocean is also warmer during
the MWP, but the magnitude is weaker than that over
the land, so the land–sea thermal contrast still increases.
The colder land–warmer ocean structure is evident in
the troposphere during the LIA (Fig. 10b), with colder
anomalies of 20.28C over land extending from approx-
imately 308–608N and warmer anomalies of 0.28C over
ocean extending from 08 to 308N.
Following the changes of temperature gradients, anom-
alous descending and ascending motion dominate the
northern land and southern ocean areas, respectively, with
anomalous low-level northerly winds between 208 to 508N.
The strongest cooling center is around the 300-hPa level
with a magnitude of 20.28C. The cooling center results
in a southward shift of high-level subtropical westerly jet
while enhancing the anomalous low-level northerlies,
similar to what happened in the later twentieth century
associated with a weakened EASM (Yu et al. 2004; Yu
and Zhou 2007). In addition, the change of meridional
temperature gradient is also consistent with the changes
of subtropical westerly jet shown in Fig. 6, a poleward
increase (decrease) of tropospheric temperature is fol-
lowed by a weakened (intensified) westerly, based on
the principle of thermal wind balance (Zhang et al.
2006).
In summary, the spatial structure of land–sea thermal
contrast and the associated circulation changes reason-
ably explains an enhanced (weakened) EASM during
the MWP (LIA), which is further confirmed by the dif-
ferences between MWP and LIA (Fig. 11), namely that
the stronger EASM during MWP relative to LIA is
dominated by the enhanced land–sea thermal contrast.
The spatial pattern of MWP–LIA tropospheric mean
temperature differences reveals warmer anomalies with
an amplitude up to 0.458Cover EAand colder anomalies
with a central magnitude of 20.58C in the tropical west-
ern Pacific and extratropical North Pacific. Following the
land–sea thermal contrast change, the southerlies pene-
trate northward into higher latitudes in the MWP than in
the LIA.
4. Summary and discussion
a. Summary
The EASM changes and the corresponding rainfall
patterns over EA during the MWP and LIA are ana-
lyzed by using the output of the MPI Earth System
Model. The association between the EASM and land–
sea thermal contrast changes is studied. Themain results
are summarized below:
1) The EASM during the MWP is stronger than that
during the LIA, as the proxy data indicated. Follow-
ing the intensified (weakened) EASM during the
MWP (LIA), the summer precipitation over eastern
China exhibits a meridional tripolar pattern. Exces-
sive (deficient) rainfall is found over northern China
but deficient (excessive) rainfall along the Yangtze
River valley during the MWP (LIA). Both similari-
ties and disparities between our model results and
the available estimates from the proxy data have been
compared: reconstructions from Chinese historical
documents and some geological evidence support our
results, but reconstructions from the Wangxiang re-
cord show disparities with the model results.
FIG. 10. Latitude–height cross section of JJA temperature av-
eraged over 1058–1228E (8C) for (a) theMWPand (b) the LIA. The
anomalies are calculated relative to themillennial mean value. The
black stippled regions denote areas that are statistically significant
at the 5% level by using a Student’s t test.
15 NOVEMBER 2012 MAN ET AL . 7861
Page 11
2) The northward water vapor transport has enhanced
(reduced) during theMWP (LIA), which leads to less
(more) moisture convergence and thus deficient
(excessive) rainfall along the Yangtze River valley
but excessive (deficient) rainfall in northern China.
Both the changes of the subtropicalwesterly jet stream
in the upper troposphere and theWPSH in themiddle
and lower troposphere are consistent with the merid-
ional shift of themonsoon rainbelt during both epochs.
A stronger (weaker) WPSH along with a northward
(southward) shift of the subtropical westerly jet stream
is evident in the warm (cold) period.
3) The meridional monsoon circulation changes show
anomalous ascending (descending) and descending
(ascending) motion respectively over the northern
land and southern ocean areas during theMWP (LIA).
An anomalous southerly (northerly) wind is seen
between 208 and 508N, which corresponds to an en-
hanced (weakened) summer monsoon.
4) The land–sea thermal contrast changes caused by the
effective radiative forcing lead to the MWP and LIA
monsoon changes. The EASM is dominated by both
the zonal thermal contrast between EA and the
North Pacific and the meridional thermal contrast
between EA and the tropical western Pacific. The
‘‘warmer land–colder ocean’’ anomaly pattern dur-
ing the MWP favors a strong EASM, while the
‘‘colder land–warmer ocean’’ anomaly pattern dur-
ing the LIA favors a weak EASM.
b. Discussion
Based on the output of MPI Earth System Model
millennial climate simulations, our analysis shows that
the EASM during the MWP is stronger than that dur-
ing the LIA. There is excessive (deficient) rainfall in
northern China but deficient (excessive) rainfall along
the Yangtze River valley during the MWP (LIA). We
compared our results with the available estimates de-
rived from the proxy data. Reconstructions from Chi-
nese historical documents and some geological evidence
support our results, but reconstructions from the Wan-
xiang record show disparities with the model results.
The reason deserves further investigation.
The model output analyzed in Liu et al. (2011) was
obtained from the millennium integrations of the cou-
pled ECHO-G model, which consists of the spectral at-
mospheric model ECHAM4 and the global ocean
circulationmodelHOPE-G. The simulationwas forced by
three external forcing factors: solar variability (Crowley
2000), greenhouse gas concentrations in the atmosphere
including CO2 (Etheridge et al. 1996) and CH4 (Blunier
et al. 1995), and an estimated radiative effect of volcanic
aerosols (Robock and Free 1996). Note that the level of
solar irradiance used to drive themodel in Liu et al. (2011)
exhibits larger change in comparison to the ‘‘state of the
art’’ estimates for solar variability applied in this study.
Both our result and the work by Liu et al. (2011) indicate
that centennial variation of the EASM is essentially a
forced response to the external radiative forcing. Besides
the similarities, however, there are also disparities be-
tween the two simulations. The results from our study
indicate that the summer precipitation over eastern China
exhibits a meridional tripolar pattern, which is charac-
terized by an out-of-phase relationship between the sub-
tropical and extratropical rainfall. This feature is consistent
with the observed structure on the interannual time scale.
The subtropical and extratropical rainfall increases si-
multaneously in Liu et al. (2011). This feature implies
that the forced mode is characterized by an in-phase
relationship between the subtropical and extratropical
rainfall. The main discrepancy in the rainfall between our
study and the work by Liu et al. (2011) is in the Yangtze
River valley whereas the extratropical (corresponding
to Northern China) response is consistent. Since the
warming in high latitudes is much larger than that over
the tropical regions during theMWP, both of themodels
FIG. 11. The differences between MWP and LIA (MWP minus
LIA) for (a) JJA mean precipitation (colored shading; mm day21)
and 850-hPa winds (vectors; m s21) and (b) JJA mean upper-tro-
pospheric (500–200 hPa) temperature (8C). The black stippled
regions denote areas that are statistically significant at the 5% level
by using a Student’s t test.
7862 JOURNAL OF CL IMATE VOLUME 25
Page 12
show enhanced southerly winds across the subtropical
and extratropical monsoon regions, indicating that the
responses of the circulations within the two models are
similar. However, the precipitation could differ signifi-
cantly even though the atmospheric circulations are the
same between different models. We suggest that the dis-
parities of the precipitation in the two simulations could
be model dependent; thus, multimodel intercomparisons
are needed in future studies. We should note that the
subtropical region (218–358N, 1008–1208E) in Liu et al.
(2011) covers southern China and the Yangtze River
valley, and excessive (deficient) rainfall is found over
southern China but deficient (excessive) rainfall along
the Yangtze River valley during the MWP (LIA) in our
simulation. If we calculate the regional-mean precip-
itation value for the subtropical region defined by Liu
et al. (2011), we also find that subtropical precipitation
was strong during MWP and weak during LIA.
We use the model output from the full-forcing exper-
iments in this study, so it is difficult to assess responses
from one specific forcing component. However, exami-
nation of the specific response fromone individual forcing
is quite important; for example, the land-use forcing could
be an important aspect for regional climate during the last
millennium. By using an atmospheric general circulation
model, Takata et al. (2009) suggested that the land cover/
use change between 1700 and 1850 could result in the
weakening of theAsian summermonsoon through changes
in the energy and water balance at Earth’s surface. Thus,
the effect of land-use forcing on the regional climate de-
serves further study by experiments in which just one
forcing component is applied.
A map with the change in land use between the MWP
and LIA is further provided in order to discuss possible
local reinforcement of the land–sea contrast due to land use
between the two periods (Fig. 12). Large areas of crop-
land were deduced for China during theMWP (Fig. 12a).
The spread of crops further developed during the LIA
compared with the MWP (Fig. 12b). The distribution
of pasture was quite different from that of cropland
(Figs. 12c,d). The pasture lands during both periods
were mainly located in Mongolia and Tibet, where herd-
ing was the traditional form of agriculture. Many parts of
eastern China showed little pasture area during those two
FIG. 12. Historical cropland and pasture area during the MWP and LIA: (a) cropland for the MWP, (b) cropland
for the LIA, (c) pasture for the MWP, and (d) pasture for the LIA. Units are percentage of grid cell. Values less than
1% are white.
15 NOVEMBER 2012 MAN ET AL . 7863
Page 13
periods. The total areas of cropland and pasture were
larger during theMWP than that during the LIA over this
region. The land-use change would favor a land–sea
thermal contrast change between the two periods.
We state the major features of symmetries between
the patterns during theMWPandLIA, and it is interesting
to see that it is not entirely symmetrical. The anomaly
of the effective radiative forcing is 20.03 W m22 for the
MWP and 20.26 W m22 for the LIA compared to the
long-term mean value, which is not symmetrical in am-
plitude between MWP and LIA. The asymmetry of the
effective radiative forcing between MWP and LIA is
consistent with the asymmetrical responses between the
two typical periods. The possible reasons for the asym-
metry in the patterns are both important and interesting.
The asymmetry between theMWPandLIA in the rainfall
pattern displays as significant negative–positive–negative
anomaly pattern extending from northwestern China to
the East Asian marginal continents. Corresponding to the
rainfall pattern, a wave train structure is evident with neg-
ative and positive anomalies by turns in the geopotential
height at 200 hPa (not shown). The wave structure of the
200-hPa geopotential height is consistent with the asym-
metry between the MWP and LIA in the rainfall pattern.
The asymmetry of the patterns and possible reasons for
the asymmetry still deserve further diagnosis in our fu-
ture study.
Additionally, our study shows evidence that the changes
of land–sea thermal contrast associated with the effec-
tive radiative forcing dominate the EASM response.
Previous studies suggest that the direct radiative effect
of solar forcing variations on the monsoon change is re-
latively weak and that dynamical responses may be more
important (Fan et al. 2009). Major climate modes during
different epochs, such as ENSO and the Pacific decadal
oscillation (PDO), may also contribute to the changes of
EASM and deserves further study. In addition, the cur-
rent diagnosis is basedon theMPImodel driven by a solar
irradiance reconstruction of weaker variability, and thus
the forcing of effective solar irradiance to EASM varia-
tion on decadal–centennial time scales may be under-
estimated. To account for uncertainty in the solar forcing,
another set of three-member ensemble simulations (E2)
with an alternative solar irradiance reconstruction of
stronger variability has been done inMPI and the results
will be used in our future diagnosis.
Furthermore, the EASM changes during the MWP
and LIA are primarily dominated by the natural vari-
ability, such as solar and volcanic forcing variability dur-
ing the last millennium. However, the future monsoon
variations could be affected by human activities, such as
anthropogenic forcings. Further, the EASM was stronger
during the MWP, whereas the monsoon has weakened
during the latter half of the twentieth century when the
warming was rapid. Further study is necessary to un-
derstand the reasons behind the EASM changes under
two similar climate backgrounds. Thus, it is not appro-
priate to suggest that the monsoon variability during the
last millennium is a possible analog for future monsoon
changes.
Acknowledgments. This work was jointly supported
by the National Natural Science Foundation of China
under Grants 40890054 and 41125017 and the Visiting
Student Program of the International Max Planck Re-
search School on Earth SystemModelling (IMPRS-ESM)
at theMax Planck Institute forMeteorology (MPI-M) in
Hamburg, Germany. The IMPRS-ESMVisiting Student
Program is funded by the ZEIT Foundation ‘‘Ebelin and
Gerd Bucerius,’’ Hamburg. We thank Drs. Chao Li and
Hongmei Li at MPI-M for the helpful comments and
discussion.
REFERENCES
Ammann, C. M., F. Joos, D. S. Schimel, B. L. Otto-Bliesner, and
R. A. Tomas, 2007: Solar influence on climate during the past
millennium: Results from transient simulations with the
NCAR Climate System Model. Proc. Natl. Acad. Sci. USA,
104, 3713–3718.Blunier, T., J. Chappellaz, J. Schwander, B. Stauffer, andD.Raynaud,
1995: Variations in atmospheric methane concentration during
the Holocene epoch. Nature, 374, 46–49.
Cao, J., S. Wang, J. Shen, and Z. Zhang, 2004: The paleoclimate
changes during the past millennium inferred from the Lacus-
trine Core in Daihai Lake, Inner Mongolia (in Chinese). Sci.
Geogr. Sin., 20, 391–396.Chang, C. P., Y. Zhang, and T. Li, 2000a: Interannual and inter-
decadal variations of the East Asian summer monsoon and
tropical Pacific SSTs. Part I: Roles of the subtropical ridge.
J. Climate, 13, 4310–4325.——, ——, and ——, 2000b: Interannual and interdecadal varia-
tions of the East Asian summer monsoon and tropical Pacific
SSTs. Part II: Meridional structure of the monsoon. J. Climate,
13, 4326–4340.Chen, H., T. Zhou, R. B. Neale, X. Wu, and G. Zhang, 2010:
Performance of the new NCAR CAM3.5 in East Asian sum-
mer monsoon simulations: Sensitivity to modifications of the
convection scheme. J. Climate, 23, 3657–3675.
Chen, Q., W. Li, and J. Mao, 1964: The mean flow fields and me-
ridional cells in the regions of southwesterly monsoon over
Southeast Asia and trade wind over the Pacific (in Chinese).
Acta Meteor. Sin., 34, 51–61.
CMA, 1981: Yearly charts of dryness/wetness in China for the last
500-year period (in Chinese). Chinese Meteorological Ad-
ministration, 332 pp.
Crowley, T. J., 2000: Causes of climate change over the past 1000
years. Science, 289, 270–277.
——, G. Zielinski, B. Vinther, R. Udisti, K. Kreutz, J. Cole-Dai,
and E. Castellano, 2008: Volcanism and the Little Ice Age.
PAGES Newsletter, No. 16, PAGES International Project
Office, Bern, Switzerland, 22–23.
7864 JOURNAL OF CL IMATE VOLUME 25
Page 14
Ding, Y., 1992: Summermonsoon rainfalls in China. J.Meteor. Soc.
Japan, 70 (1B), 373–396.
——, 1994: Monsoons over China. Kluwer, 90 pp.
Etheridge, D. M., L. P. Steele, R. L. Langenfelds, R. J. Francey,
J.-M. Barnola, and V. I. Morgan, 1996: Natural and anthropo-
genic changes in atmospheric CO2 over the last 1000 years from
air in Antarctic ice and firn. J. Geophys. Res., 101 (D2), 4115–4128.
Fan, F., M. E. Mann, and C. M. Ammann, 2009: Understanding
changes in the Asian summer monsoon over the past millen-
nium: Insights from a long-term coupled model simulation.
J. Climate, 22, 1736–1748.
Gerber, S., F. Joos, P. Brugger, T. F. Stocker, M. E. Mann, S. Stich,
and M. Scholze, 2003: Constraining temperature variations
over the last millennium by comparing simulated and ob-
served atmospheric CO2. Climate Dyn., 20, 281–299.Gonzalez-Rouco, F., H. Von Storch, and E. Zorita, 2003: Deep soil
temperature as proxy for surface air-temperature in a coupled
model simulation of the last thousand years. Geophys. Res.
Lett., 30, 2116, doi:10.1029/2003GL018264.
Goosse, H., H. Renssen, A. Timmermann, and R. S. Bradley, 2005:
Internal and forced climate variability during the last millen-
nium: A model–data comparison using ensemble simulations.
Quat. Sci. Rev., 24, 1345–1360.Hibler, W. D., 1979: A dynamic thermodynamic sea ice model.
J. Phys. Oceanogr., 9, 815–846.
Holzhauser, H., M. Magny, and H. J. Zumbuhl, 2005: Glacier and
lake-level variations in west-central Europe over the last 3500
years. Holocene, 15, 789–801.
Jungclaus, J. H., and Coauthors, 2010: Climate and carbon-cycle
variability over the last millennium. Climate Past, 6, 723–737.Kanamitsu, M., W. Ebisuzaki, J. Woollen, S. K. Yang, J. J. Hnilo,
M. Fiorino, and G. L. Potter, 2002: NCEP-DOE AMIP-II
reanalysis (R-2). Bull. Amer. Meteor. Soc., 83, 1631–1643.
Lau, K., C. Chang, and S. Shen, 1988: Seasonal and intraseasonal
climatology of summermonsoon rainfall over East Asia.Mon.
Wea. Rev., 116, 18–37.
Li, C., andM. Yanai, 1996: The onset and interannual variability of
the Asian summer monsoon in relation to land–sea thermal
contrast. J. Climate, 9, 358–375.
——, J. Wang, S. Lin, and H. Cho, 2004: The relationship between
Asian summer monsoon activity and northward jump of the
upper westerly jet location (in Chinese). Chin. J. Atmos. Sci.,
28, 641–658.
Li, H., A. Dai, T. Zhou, and J. Lu, 2010: Responses of East Asian
summer monsoon to historical SST and atmospheric forcing
during 1950–2000. Climate Dyn., 34, 501–514, doi:10.1007/
s00382-008-0482-7.
Li, S., D. Li, P. Zhao, and G. Zhang, 2009: The climatic charac-
teristics of vapor transportation in rainy season of the origin
area of three rivers in Qinhai-Xizang Plateau (in Chinese).
Acta Meteor. Sin., 67, 591–598.
Liang, X., and W. Wang, 1998: Associations between China mon-
soon rainfall and tropospheric jets.Quart. J. Roy.Meteor. Soc.,
124, 2597–2623.
Liu, J., B. Wang, H. Wang, X. Kuang, and R. Ti, 2011: Forced re-
sponse of the East Asian summer rainfall over the past mil-
lennium: Results from a coupled model simulation. Climate
Dyn., 36, 323–336, doi:10.1007/s00382-009-0693-6.
MacFarling Meure, C., D. Etheridge, C. Trudinger, P. Steele,
R. Langenfelds, T. van Ommen, A. Smith, and J. Elkins, 2006:
Law Dome CO2, CH4 and N2O ice core records extended to
2000 years BP. Geophys. Res. Lett., 33, L14810, doi:10.1029/
2006GL026152.
Mann, M. E., M. A. Cane, S. E. Zebiak, and A. Clement, 2005:
Volcanic and solar forcing of the tropical Pacific over the past
1000 years. J. Climate, 18, 447–456.
Marsland, S. J., H. Haak, J. H. Jungclaus, M. Latif, and F. Roske,
2003: The Max-Planck-Institute global ocean/sea ice model
with orthogonal curvilinear coordinates.OceanModell., 5, 91–
127, doi:10.1016/S1463-5003(02)00015-X.
Peng, Y. B., Y. Xu, and L. Y. Jin, 2009: Climate changes over
eastern China during last millennium in simulations and re-
constructions. Quat. Int., 208, 11–18.
Pongratz, J., C. H. Reick, T. Raddatz, and M. Claussen, 2008: A
reconstruction of global agricultural areas and land cover for
the last millennium.Global Biogeochem. Cycles, 22, GB3018,
doi:10.1029/2007GB003153.
Qian, W., Q. Hu, Y. Zhu, and D.-K. Lee, 2003: Centennial-scale
dry–wet variations in East Asia. Climate Dyn., 21, 77–89.
Raddatz, T. J., and Coauthors, 2007: Will the tropical land bio-
sphere dominate the climate-carbon feedback during the
twenty-first century? Climate Dyn., 29, 565–575.Ren, G., and L. Zhang, 1996: Climate of the Medieval Warm Pe-
riod in Maili Region as inferred from fossil pollen record (in
Chinese). Climatic Environ. Res., 1, 81–86.
Robock, A., and M. Free, 1996: The volcanic record in ice cores
for the past 2000 years. Climatic Variation and Forcing
Mechanisms of the Last 2000 Years, P. Jones, R. Bradley, and
J. Jouzel, Eds., Springer, 533–546.
Roeckner, E., and Coauthors, 2003: The atmosphere general cir-
culation model ECHAM5. Part I: Model description. Max
Planck Institute for Meteorology Rep. 349, 127 pp.
Schmidt, G., and Coauthors, 2011: Climate forcing reconstructions
for use in PMIP simulations of the last millennium (v1.0).
Geosci. Model Dev., 4, 33–45.
Servonnat, J., P. Yiou, M. Khodri, D. Swingedouw, and S. Denvil,
2010: Influence of solar variability, CO2 and orbital forcing
between 1000 and 1850 AD in the IPSLCM4 model. Climate
Past, 6, 445–460.
Shindell, D. T., G. A. Schmidt, R. L. Miller, andM. E. Mann, 2003:
Volcanic and solar forcing of climate change during the pre-
industrial era. J. Climate, 16, 4094–4107.
Stouffer, R. J., G. Hegerl, and S. Tett, 2000: A comparison of
surface air temperature variability in three 1000-yr coupled
ocean–atmosphere model integrations. J. Climate, 13, 513–
537.
Takata, K., K. Saito, and T. Yasunari, 2009: Changes in the Asian
monsoon climate during 1700–1850 included by preindustrial
cultivation. Proc. Natl. Acad. Sci. USA, 106, 9586–9589.
Tao, S., and L. Chen, 1987: A review of recent research on the East
Asian summer monsoon in China. Monsoon Meteorology,
C. P. Chang and T. N. Krishnamurti, Eds., Oxford University
Press, 60–92.
Tong, G. B., Y. Shi, R. J. Wu, X. D. Yang, and W. C. Qu, 1997:
Vegetation and climate quantitative reconstruction of Long-
gan Lake since the past 3000 years (in Chinese). Mar. Geol.
Quat. Geol., 17, 54–62.
Valcke, S., D. Caubel, and L. Terray, 2003: OASIS: Ocean At-
mosphere Sea Ice Soil user’s guide. CERFACS Tech. Rep.
TR/CMGC/03/69, 85 pp.
Viera, L. E. A., S. K. Solanki, N. A. Krivova, and I. Usoskin, 2011:
Evolution of solar irradiance during the Holocene. Astron.
Astrophys., 531, A6, doi:10.1051/0004-6361/201015843.
Wagner, S., S. Legutke, and E. Zorita, 2005: European winter tem-
perature variability in a long coupled model simulation: The
contribution of ocean dynamics. Climate Dyn., 25, 37–50.
15 NOVEMBER 2012 MAN ET AL . 7865
Page 15
Wang, B., 2006: The Asian Monsoon. Springer/Praxis, 787 pp.
——, R. Wu, and X. Fu, 2000: Pacific–East Asian teleconnection:
How does ENSO affect East Asian climate? J. Climate, 13,
1517–1536.
Wang, S., Z. Zhao, and Z. Chen, 1981: Reconstruction of the
summer rainfall regime for the last 500 years in China (in
Chinese). Geophys. J., 52, 117–122.
——, ——, ——, and Z. Tang, 1987: Drought/flood variations for
the last two thousand years in China and comparison with
global climatic change (in Chinese). The Climate of China and
Global Climate, D. Ye et al., Eds., China Ocean Press, 20–29.
Wang, Y. J., and Coauthors, 2005: The Holocene Asian monsoon:
Links to solar changes and North Atlantic climate. Science,
308, 854–857.
Wetherald, R. T., and S. Manabe, 1998: Cloud feedback processes
in a general circulation model. J. Atmos. Sci., 45, 1397–1416.
Wetzel, P., E.Maier-Reimer,M.Botzet, J. Jungclaus, N.Keenlyside,
and M. Latif, 2006: Effects of ocean biology on the penetra-
tive radiation in a coupled climate model. J. Climate, 19, 3973–3987.
Wu, B., T. Zhou, and T. Li, 2009a: Seasonally evolving dominant
interannual variability modes of East Asian climate. J. Climate,
22, 2992–3005.——, ——, and ——, 2009b: Contrast of rainfall–SST relation-
ships in the western North Pacific between the ENSO-
developing and ENSO-decaying summers. J. Climate, 22,4398–4405.
Wu, J., and R. Lu, 2005: Spatial pattern and landscape charac-
teristics in Otindag sandy land during the Medieval Warm
Period (in Chinese). J. Arid Land Resour. Environ., 19, 110–113.Wu, R., Z. Hu, and B. Kirtman, 2003: Evolution of ENSO-related
rainfall anomalies in East Asia. J. Climate, 16, 3742–3758.
Xie, P., and P. A. Arkin, 1995: An intercomparison of gauge ob-
servations and satellite estimates of monthly precipitation.
J. Appl. Meteor., 34, 1143–1160.
Ye, D., andG. Yang, 1979: Meanmeridional circulations over East
Asia and the Pacific Ocean (in Chinese). Sci. Atmos. Sin., 3,299–305.
Yu, R., and T. Zhou, 2007: Seasonality and three-dimensional
structure of the interdecadal change in East Asian monsoon.
J. Climate, 20, 5344–5355.——, W. Li, X. Zhang, Y. Liu, Y. Yu, H. Liu, and T. Zhou, 2000:
Climatic features related to eastern China summer rainfalls in
the NCAR CCM3. Adv. Atmos. Sci., 17, 503–518.
——, B. Wang, and T. Zhou, 2004: Tropospheric cooling and
summer monsoon weakening trend over East Asia. Geophys.
Res. Lett., 31, L22212, doi:10.1029/2004GL021270.
Zhang, D., R. Blender, X. Zhu, and K. Fraedrich, 2011: Temper-
ature variability in China in an ensemble simulation for the
last 1,200 years. Theor. Appl. Climatol., 103, 387–399.
Zhang, P., and Coauthors, 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, Y., X. Kuang, W. Guo, and T. Zhou, 2006: Seasonal evolu-
tion of the upper-tropospheric westerly jet core over East Asia.
Geophys. Res. Lett., 33, L11708, doi:10.1029/2006GL026377.
Zheng, J., W. Wang, Q. Ge, Z. Man, and P. Zhang, 2006: Pre-
cipitation variability and extreme events in eastern China
during the past 1500 years. Terr. Atmos. Ocean. Sci., 17, 579–592.
Zhou, T., and Z. Li, 2002: Simulation of the East Asian summer
monsoon by using a variable resolution atmospheric GCM.
Climate Dyn., 19, 167–180.——, and R. Yu, 2005: Atmospheric water vapor transport associ-
ated with typical anomalous summer rainfall patterns in China.
J. Geophys. Res., 110, D08104, doi:10.1029/2004JD005413.
——, and ——, 2006: Twentieth-century surface air temperature
over China and the globe simulated by coupled climate
models. J. Climate, 19, 5843–5858.
——, and L. Zou, 2010: Understanding the predictability of East
Asian summer monsoon from the perspective of land–sea
thermal contrast change in AMIP-type simulation. J. Climate,
23, 6009–6026.
——, D. Gong, J. Li, and B. Li, 2009a: Detecting and understand-
ing the multi-decadal variability of the East Asian summer
monsoon—Recent progress and state of affairs.Meteor. Z., 18,
455–467.
——, B. Wu, and B. Wang, 2009b: How well do atmospheric gen-
eral circulation models capture the leading modes of the in-
terannual variability of the Asian–Australian monsoon?
J. Climate, 22, 1159–1173.——, and Coauthors, 2009c: Why the western Pacific subtropical
high has extendedwestward since the late 1970s. J. Climate, 22,
2199–2215.
Zorita, E., J. F. Gonzalez-Rouco, H. von Storch, J. P. Montavez,
and F. Valero, 2005: Natural and anthropogenic modes of
surface temperature variations in the last thousand years.
Geophys. Res. Lett., 32, L08707, doi:10.1029/2004GL021563.
7866 JOURNAL OF CL IMATE VOLUME 25