A remarkable climate warming hiatus over Northeast China since 1998

Abstract

Characteristics and causes of global warming hiatus (GWH) phenomenon have received much attention in recent years. Monthly mean data of land surface air maximum temperature (Tmax), minimum temperature (Tmin), and mean temperature (Tmean) of 118 national stations since 1951 in Northeast China are used in this paper to analyze the changes of land surface air temperature in recent 64 years with an emphasis on the GWH period. The results show that (1) from 1951 to 2014, the warming trends of Tmax, Tmin, and Tmean are 0.20, 0.42, and 0.34 °C/decade respectively for the whole area, with the warming rate of Tmin about two times of Tmax, and the upward trend of Tmean obviously higher than mainland China and global averages; (2) in the period 1998–2014, the annual mean temperature consistently exhibits a cooling phenomenon in Northeast China, and the trends of Tmax, Tmin, and Tmean are −0.36, −0.14, and −0.28 °C/decade respectively; (3) in the GWH period, seasonal mean cooling mainly occurs in northern winter (DJF) and spring (MAM), but northern summer (JJA) and autumn (SON) still experience a warming, implying that the annual mean temperature decrease is controlled by the remarkable cooling of winter and spring; (4) compared to the global and mainland China averages, the hiatus phenomenon is more evident in Northeast China, and the cooling trends are more obvious in the cold season; (5) the Northeast China cooling trend occurs under the circulation background of the negative phase Arctic Oscillation (AO), and it is also closely related to strengthening of the Siberia High (SH) and the East Asian Trough (EAT), and the stronger East Asian winter monsoon (EAWM) over the GWH period.

Full text

1 23
Theoretical and Applied Climatology
ISSN 0177-798X
Volume 133
Combined 1-2
Theor Appl Climatol (2018) 133:579-594
DOI 10.1007/s00704-017-2205-7
A remarkable climate warming hiatus over
Northeast China since 1998
Xiubao Sun, Guoyu Ren, Yuyu Ren,
Yihe Fang, Yulian Liu, Xiaoying Xue &
Panfeng Zhang

1 23
Your article is protected by copyright and
all rights are held exclusively by Springer-
Verlag GmbH Austria. This e-offprint is for
personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.

ORIGINAL PAPER
A remarkable climate warming hiatus over Northeast
China since 1998
Xiubao Sun
1,2,3
&Guoyu Ren
2,3
&Yuyu Ren
3
&Yihe Fang
4
&Yulian Liu
5
&Xiaoying Xue
2
&
Panfeng Zhang
2
Received: 30 November 2016 / Accepted: 18 June 2017 /Published online: 5 July 2017
#Springer-Verlag GmbH Austria 2017
Abstract Characteristics and causes of global warming
hiatus (GWH) phenomenon have received much atten-
tion in recent years. Monthly mean data of land surface
air maximum temperature (Tmax), minimum temperature
(Tmin), and mean temperature (Tmean) of 118 national
stations since 1951 in Northeast China are used in this
paper to analyze the changes of land surface air temper-
ature in recent 64 years with an emphasis on the GWH
period. The results show that (1) from 1951 to 2014,
the warming trends of Tmax, Tmin, and Tmean are
0.20, 0.42, and 0.34 °C/decade respectively for the
whole area, with the warming rate of Tmin about two
times of Tmax, and the upward trend of Tmean obvi-
ously higher than mainland China and global averages;
(2) in the period 1998–2014, the annual mean tempera-
ture consistently exhibits a cooling phenomenon in
Northeast China, and the trends of Tmax, Tmin, and
Tmean are −0.36, −0.14, and −0.28 °C/decade respec-
tively; (3) in the GWH period, seasonal mean cooling
mainly occurs in northern winter (DJF) and spring
(MAM), but northern summer (JJA) and autumn
(SON) still experience a warming, implying that the
annual mean temperature decrease is controlled by the
remarkable cooling of winter and spring; (4) compared
to the global and mainland China averages, the hiatus
phenomenon is more evident in Northeast China, and
the cooling trends are more obvious in the cold season;
(5) the Northeast China cooling trend occurs under the
circulation background of the negative phase Arctic
Oscillation (AO), and it is also closely related to
strengthening of the Siberia High (SH) and the East
Asian Trough (EAT), and the stronger East Asian winter
monsoon (EAWM) over the GWH period.
1 Introduction
Intergovernmental Panel on Climate Change (IPCC) fifth as-
sessment report indicated that the global land surface is
warming at a rate of 0.095–0.107 °C/decade during 1901–
2012 (Stocker et al. 2013). Global warming is therefore a
consensus in climatological community. However, Cater
(2006) found that global warming appears to stop or slow
down since 1998. Easterling and Wehner (2009) analyzed
the observational data and indicated that the global land sur-
face air temperature is not significantly warming as expected
during the last decade, with some areas even cooling, and this
phenomenon is called global warming hiatus (GWH). The
GWH, in both global and regional scales, has received a wide-
spread concern in the last years (e.g., Kerr 2009; Franzke
2014; Fyfe and Gillett 2014;Lietal.2015;Anetal.2016).
By analyzing the seasonalcharacteristics of the GWH, Kosaka
and Xie (2013) and Trenberth et al. (2014b) found that the
warming trends in the northern hemisphere significantly slow-
down in winter but continually increase in summer. A recent
analysis showed that the GWH mainly appears in the low and
middle latitudes of global lands, with surface air temperature
*Guoyu Ren
guoyoo@cma.gov.cn
1
College of Atmospheric Science, Nanjing University of Information
Science &Technology, Nanjing 210044, China
2
Department of Atmospheric Science, School of Environmental
Studies, China University of Geosciences, Wuhan 430074, China
3
Laboratory for Climate Studies, National Climate Center, China
Meteorological Administration, Beijing 100081, China
4
Liaoning Meteorological Bureau, Shenyang 110001, China
5
Heilongjiang Meteorological Bureau, Haerbin 150001, China
Theor Appl Climatol (2018) 133:579–594
DOI 10.1007/s00704-017-2205-7
Author's personal copy

at most stations there witnessing a stable or decreased trend
after 1998 (Sun et al. 2017). However, further studies revealed
that the GWH seems not detectable in change of global ex-
treme temperature event frequency all the time (Seneviratne
et al. 2014). Karl et al. (2015), based on a new land and ocean
dataset, recently also indicated that global surface and land
surface warming had never slowed down, challenging the ex-
istence of the GWH.
In recent years, however, more researchers confirmed
the GWH phenomenon on global and regional scales, and
many works were focused on the possible reasons of the
GWH (Fyfe et al. 2016). A number of studies indicated
that the mechanisms of the GWH may be related to the
external forcing and natural variability. The influence
from external forcing may include the prolonged solar
minimum (Hansen et al. 2011), the increased volcanic
eruptions (Balmaseda et al. 2013;Santeretal.2014),
the reduced water vapor in stratospheric (Solomon et al.
2010,2011), and the increased emissions of anthropogen-
ic aerosols (Lean and Rind 2009). On the other hand,
mechanisms of the natural variability proposed to explain
the GWH phenomenon include the increased ocean heat
uptake especially the layer below 700 m (e.g., Chen and
Tung 2014;Hansenetal.2011; Meehl et al. 2013;
Trenberth et al. 2014a), the cool waters of the equatorial
Pacific surface (Kosaka and Xie 2013), and the Pacific
Decadal Oscillation (PDO) into the negative phase since
1998 (Tollefson 2014). Several recent studies have attrib-
uted the GWH to the 60-year-quasi-periodic natural cli-
mate variability and have provided a theoretical frame-
work for the dynamical processes and the decadal-scale
prediction of temperature variation. Li et al. (2013a,b)
indicated that North Atlantic Oscillation (NAO) leads
the multidecadal variability in Northern Hemispheric sur-
face temperature by about 15–20 years through a delayed
effect on the North Atlantic Ocean, and the recent NAO
decadal weakening can be a useful predictor of the hiatus
in both Atlantic Multidecadal Oscillation (AMO) and
Northern Hemispheric mean surface temperature. Sun
et al. (2015) proposed a new Bdelayed oscillator theory^
of the North Atlantic decadal-scale air-sea coupling to
understand the underlying physical mechanisms of the
60-year-quasi-periodic natural climate variability.
For mainland China, annual mean temperature anoma-
lies remains at a high level since 2001, but the warming
trend has also slowed down (Tang et al. 2012). Li et al.
(2015) found that the annual mean maximum temperature
(Tmax)increasingtrendhadsloweddowninmainland
China since 1998, and although the Tmax significantly
increased in summer (JJA), the trend of minimum temper-
ature (Tmin) obviously decreased in winter (DJF). Yan
and Liu (2014) indicated that most of the Tibetan
Plateau still showed strong warming trend since 1998.
Recently, based on the ice-core record, An et al. (2016)
also founded a remarkable cooling phenomenon in north-
western Tibetan Plateau region.
Northeast China has been one of the most significant
warming regions in mainland China and East Asia for
the last more than five decades (Ren et al. 2012;Sun
et al. 2016). Investigation into the spatial and temporal
patterns of the long-term land surface air temperature
change during the GWH period in Northeast China will
help in understanding of the climate forcing mechanism
of the larger-scale GWH. Furthermore, the Northeast
China Plain is one of the most productive regions of
grain in the world. The grain yield in this region
amounts to ~20% of China’s total. Previous studies
found that the crop production has strong dependence
on the growth season mean temperature and heat condi-
tions (Lobell et al. 2011). The cooling of the warm
season will probably increase the instability of agricul-
tural production in Northeast China (Jin et al. 2002;Liu
et al. 2013a,b). It is thus also important to understand
the characteristics of annual and the warm-season tem-
perature changes in the GWH period in Northeast China,
and to predict the future change in surface air tempera-
ture and heat condition so that an adaption strategy can
be developed to reduce the risk of the summer low
temperature disasters.
In this paper, we analyze the changes in Northeast
China land surface air temperature in recent 64 years,
with an emphasis on the GWH period. The paper is orga-
nized as follows. We describe the data and methods used
in this paper in Section 2after the introduction. Results of
the analysis are presented in Section 3.Abriefanalysisof
circulation background of the warming slowdown is of-
fered in Section 4. We offer a discussion of the results in
Section 5. Finally, main conclusions are presented in
Section 6.
2Dataandmethods
2.1 Data sources and study region
The source of the monthly mean measurements used in
our current analysis is the China Homogenized Historical
Temperature Dataset (CHHTD-V1.0), including 2419 sta-
tions, with a record length of 64 years (1951–2014),
which included a quality control and homogenization pro-
cedure (Cao et al. 2016). The inhomogeneity problem can
be considered to have little influence on the dataset. The
study area comprises the east region of Inner Mongolia,
Liaoning, Heilongjiang, and Jilin provinces in Northeast
China. In this paper, we initially selected only those na-
tional stations that had at least 70% data coverage since
580 X. Sun et al.
Author's personal copy

1951 and at least 20 years of records in the base period
1961–1990 in the study region. We mainly focused on the
hiatus phenomenon, and therefore we removed the sta-
tions with records of less than 15 years in length in the
GWH period. We also considered the problem of missing
data by discarding the annual (season) value with monthly
records less than 8 (3) months in calculating annual
(seasonal) temperature anomalies. Finally, we obtained a
dataset consisting of 118 stations across the study area for
use in our analysis. There were only 26 stations in 1951,
the number increased to 70 in 1955, and substantially
increased to 118 after 1960. The study region (gray areas),
station distribution (various color points), and the alti-
tudes of the stations are shown in Fig. 1.
This study utilized six circulation indices to analyze the
abnormal characteristics of the atmospheric general circu-
lation (Section 4). The monthly mean indices were from
the National Climate Center (NCC) of the China
Meteorological Administration (CMA), which were man-
aged and regularly updated by the Climate Diagnostics
and Prediction Division of the NCC. The National
Center for Environmental Prediction/National Center for
Atmospheric Research (NCEP/NCAR) reanalysis data, in-
cluding the monthly wind, sea-level pressure (SLP), and
geopotential heights at 500 hPa and 1000 hPa since 1951
with a spatial resolution of 2.5° × 2.5°, were used for
calculating the indices.
2.2 Methods
We constructed regional series by the reference method of
Jones et al. (2001). Firstly, the study area was divided into
a total of 42 grid boxes with the spatial resolution of
2° × 2°, and each grid boxes had at least one station as
showninFig.1(gray areas). Then, gridding of the
temperature anomalies are made by averaging all values
within 2° × 2° grid boxes. Finally, the time series were
constructed by area-weight averaging all the grid boxes
with data using the cosines of the central latitudes of the
grid boxes as weight coefficients.
The analysis used 1961–1990 as the base period,
mainly because of the better spatial coverage of stations
in the period, and the comparability with previous stud-
ies for surface air temperature. China lies in the north-
ern hemisphere, and therefore seasons were divided into
spring (MAM), summer (JJA), autumn (SON) and win-
ter (DJF, December to February of the following year).
Annual mean values were those from January to
December.
The linear trends of the anomaly series were obtained by
using least squares method to calculate the linear regression
coefficients between temperature and ordinal numbers of time
(e.g., i=1,2,3…64 for 1951–2014). The significance of the
linear trends of temperature series was judged by using two-
tailed student’sttest method. In this study, a trend was con-
sidered to be statistically significant if it is significant at the
5% (P< 0.05) level. We also used a nonparametric
Kendall’staubasedSen’s slope estimator (Sen 1968)to
calculate trends, since this method does not assume a dis-
tribution for the residuals and is robust to the effect of
outliersintheseries,andithasbeenwidelyusedinthe
studies of hydrological and extreme climate change (e.g.,
Alexander et al. 2006;Zhaoetal.2016; Kosaka and Xie
2013). The significance of the Sen’s slope is judged by
using Mann-Kendall test method (Kosaka and Xie 2013;
Zhao et al. 2016).
In Section 4, empirical orthogonal function (EOF)
(Lorenz 1956) and singular value decomposition (SVD)
method (Golub and Reinsch 1970; Lathauwer et al.
2000) were used to analyze the possible influence of the
Fig. 1 Map of study region (gray
shade in bottom right corner)and
location of 118 stations (various
color points). The elevation of
stations is indicated by the
different colors. The grid boxes at
a spatial resolution of 2° × 2°used
for the estimation of regional
average are shown by the gray
areas
A remarkable climate warming hiatus 581
Author's personal copy

circulation factors on the observed temperature change in
Northeast China.
2.3 Circulation indices
We used six circulation indices in Section 4, and their defini-
tions are as follows:
Arctic Oscillation (AO) index (I
AO
): normalized time
coefficient series of the EOF1 of 1000 hPa height
anomaly field (relative to the 1981–2010) (20–90° N)
(Thompson and Wallace 1998).
Siberia High (SH) index (I
SH
): winter mean sea level pres-
sure anomaly (relative to 1981–2010) in SH region (45–70°
N, 80–110° E). The index is the normalized value.
East Asian Trough (EAT) index (I
EAT
): difference between
the maximum height and the minimum height on the 500 hPa
geopotential heights in the area of 110–170° E and 30–55° N.
Station of the East Asian trough (I
EATS
): the mean position
of the EAT line on the 500 hPa geopotential heights in the area
of 110–170° E and 30–55° N.
East Asian winter monsoon (EAWM) strength index
(I
EAWM
)(Zhuetal.2008):
IEAWM ¼U500 25-−35-N;80-−120-EðÞ
−U500 50-−60-N;80-−120-EðÞ
ð1Þ
where U500 25-−35-N;80-−120-EðÞ
and U500 50-−60-N;80-−120-EðÞ
are
the mean values of 500 hPa zonal wind in their respective
regions. The index is the normalized value.
Meridional index over Asia (I
M
): we firstly divided the East
Asian region (45–65° N, 60–150° E) into three areas (60–90°
E, 90–120° E, 120–150° E), and then calculated I
M
for each
area according to formula (2):
Fig. 2 Linear trends of
temperature series for different
periods (1951–2014, 1951–1997,
and 1998–2014). Figure on the
left shows the annual mean time
series (°C, black lines)ofTmax
(a), Tmin (b), and Tmean (c),
anomalies relative to 1961–1990.
Figure on the right shows the
linear trends for the period 1951–
2014 (°C/decade, red bars),
1951–1997 (°C/decade, orange
bars), and 1998–2014 (°C/
10 years, blue bars). Error bars
indicate two times the standard
deviation. Statistically significant
(P< 0.05) trends are marked with
asterisks.Figureonthebottom
shows the interdecadal mean
temperature anomalies of Tmax
(red lines), Tmin (blue lines), and
Tmean (black lines)(d)
582 X. Sun et al.
Author's personal copy

IM¼1
n∑
n
j¼1
1
cosφ
∂z
∂λ

j

¼1
n∑
n
j¼1
1
m∑
m
i¼1
1
cosφj
Δzi
Δλ
!
j

¼1
mnΔλ ∑
n
j¼1
∑
m
i¼1
Δzi
cosφi

j

ð2Þ
where nis the number of partitions within the study region; m
is the number of φ(here m=3);Δλ= 15° longitude; φ
1
,φ
2
,
and φ
3
are 45° N, 55° N, and 65°N respectively; and Δz
i
is the
difference between the height of every 15° (Δλ) longitude.
Fig. 3 Trends of Northeast China
temperature in specific periods
(1951–2014 and 1998–2014).
Figure on the left shows the
spatial distribution of the Tmax
(a), Tmin (c) and Tmean (e)
trends for 1951–201. Figure on
the right shows the spatial
distribution of the Tmax (b),
Tmin (d) and Tmean (f) trends for
1998–2014.The details above
each plot indicate the different
periods and the trends of regional
mean
A remarkable climate warming hiatus 583
Author's personal copy

3Results
3.1 Annual temperature changes
The annual mean time series of Tmax, Tmin, and mean tem-
perature (Tmean) anomalies are shown in Fig. 2a–c. For the
study period (1951–2014), annual mean temperature anoma-
lies reflect significant positive trends (P<0.05),andthe
warming rates of Tmax, Tmin, and Tmean are 0.20, 0.42,
and 0.34 °C/decade, respectively, over the study region (red
bars in Fig. 2a, b, c). The warming rate of Tmin is about twice
that of Tmax, which suggests that the warming in recent de-
cades can be attributed to stronger increases in Tmin. It is also
clear that the warming trend of Tmean in the study region is
larger than that of mainland China (Ren et al. 2005;Lietal.
2015) and global average (Jones et al. 2001). Based on the
time series, we further found that 2007 was the warmest year
in the study region since 1951. For the period 1951–1997, the
warming rates of annual temperature are larger than that in
whole study period (orange bars in Fig. 2a, b, c). The decadal
mean of temperature anomalies are shown in Fig. 2d, which
clearly indicates the inter-decadal variation in temperatures.
The decadal mean anomalies of Tmin are slightly lower than
those of Tmax before 1980, but considerably higher than
Tmax after 1980. The anomaly of Tmin clearly increased from
the late 1970s, while Tmax and Tmin significantly increased
from the late 1980s. Previous studies have indicated that sig-
nificant warming began at the majority of the stations in main-
land China in the mid-1980s (Ren et al. 2005; Tang et al.
2005). All of the above analyses reveal that the significant
warming occurred in the study region later than in most parts
of China.
In the GWH period, annual temperatures show clear nega-
tive trends (blue bars in Fig. 2a–c). The cooling trends since
1998 in Tmax, Tmin, and Tmean respectively are −0.36,
−0.14, and −0.28 °C/decade. These negative trends may be
largely attributed to several cold years around 2010, particu-
larly 2010 and 2012, as shown in Fig. 2a, b, c. Remarkably, all
of the trends in the hiatus period are not significant (P<0.05),
with a higher level of uncertainty (black error bar in Fig. 2a, b,
c), which can mainly be attributed to the short series length.
Overall, a remarkable cooling phenomenon in the GWH peri-
od occurred over the study region, and Tmax showed more
obvious cooling trends than Tmin and Tmean.
Spatial patterns of the decadal trends in temperature in the
study region are shown in Fig. 3. From 1951 to 2014 (Fig. 3a–
e), all of the grids consistently show a warming spatial pattern,
however, with different warming rates. The warming rates of
Tmin and Tmean were above 0.4 °C/decade, and the larger
change occurred in the region of north of 46° N and the
Mongolian Plateau. Although Tmax also exhibits widespread
warming, the warming rate was considerably lower than Tmin
and Tmean. The spatial pattern of Tmax indicates an increase
in the warming rate of Tmax with altitude increase. This result
is in agreement with Dong et al. (2015), who showed that the
temperature trend increased from 200 to 2000 m with altitude
increase.
In the GWH period, the majority of grid boxes exhibit
widespread cooling trends, except some grid boxes in the
north of the study region (Fig. 3b–f). Moreover, the trends
of Tmax and Tmean show a wider cooling range than Tmin.
More than ~90% of the grid boxes show decreasing trends in
Tmax and Tmean, while ~30% of the grid boxes show
warming trends of Tmin. In general, the above evidence sug-
gests that the hiatus phenomenon that has occurred since 1998
can be more strongly attributed to the substantial decreases in
Tmax than to decreases in Tmin.
3.2 Seasonal temperature changes
Decadal trends of seasonal temperature during the different
periods are shown in Fig. 4. During 1951–2014 (Fig. 4red
bar), Tmax, Tmin, and Tmean consistently show significant
warming trends (P< 0.05), and the warming rate in winter and
spring are larger than in summer and autumn. The warming
rate of Tmin was significantly higher than that of Tmax. In the
GWH period (Fig. 4, blue bar), hiatus phenomena were ob-
served in spring, summer, and particularly in winter with a
winter cooling rate lower than −0.8 °C/decade. Conversely,
the warming rate in autumn was higher relative to the recent
64 years. In general, the cooling trends of annual temperature
in Northeast China were mainly controlled by the cooling
trends in winter and spring during the GWH period, indicating
that the warm and cold seasons have become more polarized.
In addition, all the trends in the GWH period are associated
with a higher level of uncertainty (black error bar in Fig. 4).
Spatial patterns of Tmean seasonal trends in the GWH
period and the whole period are shown in Fig. 5.Becauseof
the similar spatial distribution among Tmax, Tmin, and
Fig. 4 Decadal trends (°C/decade, red and blue bars)ofTmax(left),
Tmin (middle), and Tmean (right). Statistically significant (P< 0.05)
trends are marked with asterisks.Error bars indicate two times the
standard deviation
584 X. Sun et al.
Author's personal copy

Tmean, we discuss only the spatial pattern of Tmean. From
1951 to 2014, greater warming phenomenon in spring and
winter occurred through most of the grid boxes. Conversely,
in the GWH period, almost all of the warming grid boxes in
Fig. 5 Season trends of
Northeast China Tmean change in
specific periods (1951–2014 and
1998–2014). The details above
each plot indicate the different
periods and the trends of regional
mean. Figure on the left (a, c, e, g)
shows the season trends for 1951-
2014. Figure on the right (b, d, f,
h) shows the season trends for
1998-2014
A remarkable climate warming hiatus 585
Author's personal copy

the whole period showed cooling trends in spring and winter.
The largest decrease of spring mean temperature occurred in
central and southern parts of the study region. In summer and
autumn of the GWH period, the trends increased from slight
cooling to considerable warming with increased latitude. In
autumn, the warming trends in the GWH period were consid-
erably larger than in the past 64 years, especially in the areas
north of 46° N.
3.3 Monthly temperature changes
Linear trends of monthly Tmax (a), Tmin (b), and Tmean (c)
in the whole period and the GWH period are shown in Fig. 6.
From 1951 to 2014, warming trends were observed in all
months, and the warming rates during January to June were
larger than those during July to December. February showed
the largest warming trend and the lowest change usually oc-
curred in December. During the GWH period, there were
7 months that consistently showed cooling trends, which is
more than the warming months. The larger cooling trends
mainly occurred from February to April, and the largest
warming month was November. Moreover, all of the trends
in the GWH period were associated with a high error level.
The spatial patterns of Tmean trends for February and
November in the whole period (a, c) and the GWH period
(b, d) are shown in Fig. 7. In the GWH period, the spatial
distribution of November (d) and February (b) show opposite
patterns. In February, all the grid boxes showed cooling trends
with a cooling rate lower than −0.8 °C/decade. However, in
November, all the grid boxes show greater warming trends
especially north of 46° N where the warming rates even
reached 2.0 °C/decade.
3.4 Comparison of the trends based on different trend
estimator
We used Sen’s slope and least squares estimator to assess the
linear trends of the full study period (1951–2014) and the
GWH period (1998–2014), and to compare the results based
on the two methods. Table 1shows the differences between
Sen’s slope trends and least squares trends. For the period
1951–2014, there are almost no differences on the trends ob-
tained using the two estimation methods. The trends for hiatus
period by using the Sen’s slope method are slightly lower than
that by using the least squares method, but all the results show
consistent hiatus phenomenon over Northeast China (Table 1).
For statistical significance, both of the two estimation methods
show significant long-term trends for the entire period, and
non-significant trends of the hiatus period. Therefore, the
two methods have little impact on the estimated results of
linear temperature trends and their significance over
Northeast China, but the magnitudes of the trends do have
slight differences, and this may call for a caution in estimating
the temperature trends using the least squares method.
3.5 Comparison of Northeast China, mainland China,
and global analyses
Tab le 2gives the annual mean surface temperature trends in
Northeast China (Section 3.1–3.3), mainland China (Li et al.
2015), and the global averages (Karl et al. 2015;Kosakaand
Xie 2013). Because these studies used different datasets and
base periods, it only shows an approximate comparison of the
temperature trends.
Global analysis results suggested that the annual mean tem-
perature increase was significant during 1998 to 2014 over the
Fig. 6 Monthly trends (°C/decade, red and blue bar)ofmeanTmax(a),
Tmin (b), and Tmean (c). Statistically significant (P< 0.05) trends are
marked with asterisks.Error bars indicate two times the standard
deviation
586 X. Sun et al.
Author's personal copy

global land surface (Karl et al. 2015). In Northeast China, how-
ever, the annual mean surface air temperature showed a cooling
trend over the GWH period. This illustrates that there remains a
remarkable hiatus phenomenon at the regional scale, in areas
such as Northeast China. For the seasonal hiatus, the compari-
son between Northeast China and the globe (Kosaka and Xie
2013) reveals that both consistently showed cooling trends dur-
ing DJF and MAM, and the season with the largest cooling
trends was DJF. During JJA, global and Northeast China con-
sistently showed warming trends. However, during SON, the
global result showed a slight cooling trend, but Northeast China
exhibited a greater warming trend.
Comparison of the temperature changes between mainland
China (Li et al. 2015) and the study region for annual mean
Tmax and Tmin indicated a better agreement in the GWH
period, with both showing remarkable cooling trends. For
seasonal analysis, the negative trends in Northeast China were
larger than those in mainland China during MAM. During
SON, the most significant difference between mainland
China and Northeast China was in Tmax, and there was clear
warming in Northeast China but slight cooling over mainland
China. In general, this analysis illustrates that the hiatus phe-
nomenon was more evident in Northeast China, and the
cooling trends were clearer in the cold season.
4 Circulation anomaly during the GWH
Sections 3.2 and 3.3 indicated that the decreasing trend of
annual mean temperature in Northeast China during the
GWH period is mainly controlled by the cooling of winter
and spring, especially the winter cooling. In this section, we
further analyzed the relationship between the warming hiatus
of winter over Northeast China with the atmospheric circula-
tion anomaly in larger spatial scale.
EOF method was used to analyze the winter temperature of
Northeast China during the period 1951–2014. The results
showed that the first mode contribution was 80%, and the
temperature in Northeast China region shows a spatially con-
sistent change. This result is similar with the previous research
by Fang et al. (2013). It shows that the winter temperature in
Northeast China shows the characteristics of synchronous
change.
We used SVD method to analyze the correlations of the
winter mean temperature in Northeast China with the SLP
field, the 500 hPa geopotential height field, and the 500 hPa
zonal wind field. The variance contribution of SVD for SLP
field, 500 hPa geopotential height field, and 500 hPa zonal
wind field are 91, 90, and 85% in the first mode components,
respectively, with the correlation coefficients of the
Fig. 7 Tmean trends of
February (a, b) and November (c,
d) temperature over Northeast
China in specific periods (1951–
2014 (a, c) and 1998–2014 (b, d).
The details above each plot
indicate the different periods and
the trends of regional mean
A remarkable climate warming hiatus 587
Author's personal copy

normalized time coefficient series between the three fields and
the winter temperature in Northeast China being 0.79, 0.7, and
0.78 (P< 0.01), respectively. These illustrate that there are
good relationships between winter temperature and SLP field,
500 hP a geopotential height field, and 500 hPa zonal wind
field.
The distribution map of heterogeneous correlation for SVD
first mode between the winter temperature field in Northeast
China and SLP field, 500 hPa geopotential height field (Fig.
8a, b) show that there are two significant correlation regions.
The Arctic region shows a significant negative correlation,
and the middle latitude regions show significant positive cor-
relation. The two regions are remarkably related to the con-
trolling area of AO and SH. The distribution of heterogeneous
correlation between zonal wind field and the winter tempera-
ture field (Fig. 8c) shows that the two significant correlation
regions are consistent with the regions used to define the
EAWM index by Zhu et al. (2008).
Overall, all above analyses indicate that there are good
relationships between AO, SH, EAWM, and the winter tem-
perature in Northeast China. If AO is in its positive (negative)
phaseandatthesametimeSHandEAWMarestronger
(weaker) than normal, the winter mean temperature will be
anomalously warm (cold) over Northeast China. Besides,
EAT is also an important system related to cold wave in
Northeast China and is closely related to the EAWM (Ding
et al. 2014). Several previous analyses of the EAT indicated
that its strengthening (weakening) is always accompanied by
cold (warm) winter in Northeast China (Chen et al. 2014;
Leung and Zhou 2015a). At the same time, the mean position
of the EAT can also influence the winter temperature of east-
ern China (Leung and Zhou 2015b).
Tabl e 1 Trends of temperature in different periods over Northeast China based on different trend estimation methods
Regions Periods Elements
Season
Annual
MAM JJA SON DJF
Northeast
China
1951−2014
Tmax
0.29
0.27
0.14
0.16
0.16
0.17
0.26
0.23
0.19
0.20
Tmin
0.52
0.52
0.28
0.30
0.35
0.34
0.56
0.52
0.42
0.42
Tmean
0.45
0.43
0.23
0.25
0.28
0.29
0.40
0.38
0.34
0.34
1998−2014
Tmax
–0.70
–0.75
–0.07
0.03
0.31
0.50
–1.07
–1.06
–0.39
–0.36
Tmin
–0.43
–0.37
0.33
0.30
0.26
0.50
–1.35
–0.94
–0.09
–0.14
Tmean
–0.53
–0.54
0.12
0.05
0.41
0.45
–1.30
–1.03
–0.26
–0.28
aaaaa
aaaaa
aaaaa
aaaaa
aaaaa
aaaaa
The red numbers represent Sen’s slope trends, and the blue numbers represent least squares trends (unit: °C/decade)
a
The trends are significant at the 5% level
588 X. Sun et al.
Author's personal copy

These results are generally consistent with the previous
studies (Sun and Li 2012; Li et al. (2013a,b); Ding et al.
2014; Liu et al. 2013a,b;Fangetal.2013). Fig. 9shows the
decadal mean of temperature anomalies in Northeast China
(a), mean position of the EAT (I
EATS
) (b), EAT index (I
EAT
)
(c), Meridional Index (I
M
)(d),AOindex(I
AO
)(e),SHindex
(I
SH
) (f), and EAWM index (I
EAWM
) (g) for DJF (the arrows
represent increasing or decreasing trends). It is clear that all of
the indices trends have reversed during the 1990s. During the
1990s, I
EATS
was located in the Japan east Pacific (about
148°E), and the I
EATS
and I
M
were consistently in a weak state,
limiting the cold air of the polar region from entering
Northeast China, with a result of abnormal high winter
mean temperature in the study area (Fig. 2a). Decadal
mean values of 2000s showed that I
EATS
moved west-
ward to Japan (about 144° E), closer to Northeast
China. Moreover, the I
EAT
and I
M
changed from weak
phases in the 1990s to strong phases, AO changed to a
negative phase, and SH, EAWM, EAT, and meridional
wind are all consistently in their strong phases. The
circulation abnormality benefited the more frequent
bursts of the winter cold waves and the decrease of
seasonal mean temperature in Northeast China during
the GWH period.
Tabl e 2 Comparison of Northeast China, mainland China (Li et al. 2015), and global (Karl et al. 2015; Kosaka and Xie 2013) temperature changes for
the GWH period
Sources Regions Periods Data sets Methods Elements
Season
Annual
MAM JJA SON DJF
Karl et
al. (2015)
Global
land-surf
ace
1998−2014
ISTI
Databank
v1.0.0
Sen’s
slope
Tmean
Kosaka and
Xie. (2013)
2002−2012
HadCRUT
v 4.1.1.0
Sen’s
slope
Tmean
Li et
al. (2015)
Mainland
China
1998−2012
China
Homogenized
Temperature
Data set
(CHTD)
Sen’s
slope
Tmax
Tmin
Tmean
This paper
Northeast
China
1998−2014 CHHTD-V1.0
Sen’s
slope
Tmax
Tmin
Tmean
<–0.25 <<0.0 <<0.25<
<–0.5<<0.5<°C/decade
The red (blue) arrows indicate increasing (decreasing) trends, and the number of arrows indicates the range of trend values (see the legend below the
table)
ISTI Databank v1.0.0 International Surface Temperature Initiative Databank v1.0.0, HadCRUT v 4.1.1.0 Hadley Centre-Climate Research Unit com-
bined land surface air temperature v 4.1.1.0, CHTD China Homogenized Temperature Data set (total about 860 stations in China), CHHTD-V1.0 China
Homogenized Historical Temperature Data set-V1.0 (total 2419 stations in China)
A remarkable climate warming hiatus 589
Author's personal copy

Figure 10 shows a schematic diagram of the possible mech-
anism of winter warming hiatus phenomenon in Northeast
China. The hiatus phenomenon appears under the background
of negative AO and the stronger SH, EAWM, EAT, and me-
ridional wind. Under the synergistic effects of the atmospheric
circulation factors, the cold air of the polar region more easily
moved into Northeast China, and the winter temperatures cor-
respondingly experienced the cooling phenomenon.
5Discussion
Most observational data used in this paper had a relatively
long and high-quality record, especially after 1960.
However, the station spatial coverage was poorer during
1951–1960. Therefore, the temperature anomalies calculated
for earlier period (1951–1960) have a larger uncertainty.
Furthermore, the poorer data coverage can also impact the
long-term trend estimate (Brohan et al. 2006), leading to
uncertainties of the trend values in the 1951–2014 period.
However, the uncertainties of the trend estimates are relatively
small by comparing the previous analyses of both the study
region and mainland China. In addition, this work was fo-
cused on the GWH period, and the temperature trends during
the whole time period could be regarded as a background
change. In the last 50 and 20 years, the quality and spatial
coverage of the data are better than the early data, so the bias
from data could not substantially impact the final results.
Overall, our analysis result confirms a remarkable warming
slowdown in Northeast China during the GWH period, and
the hiatus is more obvious than those reported for mainland
China and the globe. This is in a strong contrast to the fact that
Northeast China had experienced the most significant
warming in mainland China and East Asia by late 1990s
(Hansen et al. 2006; Ren et al. 2012; Jones et al. 2012). It
would be interesting to examine whether or not the other rap-
idly warming regions of the global lands also see a larger
decline of surface air temperature during the GWH period.
SLP 500 hPa geopotential heights
500 hPa zonal wind
a
c
b
Fig. 8 The distribution of heterogeneous correlation for SVD first mode
between the winter mean temperature field in Northeast China and SLP
field (a), 500 hPa geopotential heights field (b), and 500 hPa zonal wind
field (c). The shaded areas represent those with significant correlations,
and ±0.27 (±0.35) indicate that the correlations are significant at the 0.05
(0.01) level
590 X. Sun et al.
Author's personal copy

In Section 4, the direct influential factors of the Northeast
China temperature decline during the GWH period was inves-
tigated in terms of large-scale atmospheric circulation anom-
aly. We just founded out the factors of winter temperature
change in Northeast China. In reality, however, there remains
a complex physical mechanism between EAWM and the EAT
(Leung et al. 2016), and studies have also indicated that a
range of factors affecting the EAT and EAWM. Wu and
Wang ( 2002) analyzed the winter AO, SH, and EAWM and
indicated that, compared with impacts of the winter AO, the
SH showed more direct and significant impacts on the EAWM
and EAT, and this conclusion was confirmed by the other
studies (e.g., Gong et al. 2001; Li and Wu 2012; Wu and
Wan g 2002;SunandLi2012). The negative temperature
anomalies in the southern part of the EAT could also be rea-
sonably explained by the phase of the El Niño–Southern
Oscillation (ENSO) (e.g., Leung et al. 2016; Wang and He
2012;Zhouetal.2013). Besides, Clark and Serreze (2000)
found that snow depth and extents over land can affect the
Pacific Ocean temperature and thus affect the EAT and
EAWM. Recently, Qiao and Feng (2016) found that a signif-
icant positive correlation exists between the December NAO
a
b
c
d
e
f
g
Fig. 9 Decadal mean of
temperature anomalies of
Northeast China (a,°C),I
EATS
(b,
°E), I
EAT
(c), I
M
(d), I
AO
(e), I
SH
(f), and I
EAWM
(g) indices for DJF
(the arrows represent increasing
or decreasing trends)
A remarkable climate warming hiatus 591
Author's personal copy

and the following February EAT, and much of China features
cold anomalies with significant signals observed over
northeastern China in strengthened EAT and NAO years.
Huang et al. (2013) analyzed the variations of the intensity
of EAT and indicated that the variation of EAT may be related
to the eastward propagation of Rossby wave. Above studies
illustrate that the physical mechanisms for the hiatus phenom-
enon in Northeast China are complex, and the interactions
among the atmospheric, oceanic, and land surface factors
and the fundamental mechanisms still require further
examination.
Whatever the reason for the warming slowdown, the
cooling climate in Northeast China during the past 17 years
may pose a huge challenge to the agricultural activity and
natural system. Northeast China plain is one of the most pro-
ductive regions of grains in the world. The grain yield
amounts to ~20% of China’s total. The growth season here
mainly concentrates in March–October, but the cold weather
in spring and summer frequently causes loss of grain produc-
tion in the region. Our analysis in Section 3shows that the
hiatus phenomenon is more evident in Northeast China espe-
cially in the cold season (MAM and DJF). The cooling trends
in MAM signified that the accumulated temperature and heat
conditions in the growth season may have worsen during the
last two decades. It is worth noting if the cooling trend will
continue in decades to come, with Li et al. (2013a,b), Sun
et al. (2015), and others thought that it is very likely. If it will
continue, the instability of agricultural production will in-
crease in the future in this important grain production region
of mainland China. This is especially anxious considering the
widely hold expectation that climate will continue to become
warmer in the decades to come without doubt in Northeast
China.
It is also interesting to note that the phonological
rhythms of a few plants seem to adjust to changed cli-
mate condition in Northeast China (Zhao 2016). An
example comes from a species of willow and Salix
matsudana, which has exhibited a delayed dates of
leafing and flowering in spring since late 1990s (Zhao
2016). Obviously, further investigation is needed to ex-
amine the phonological and vegetation response of other
species to the recent change in surface air temperature
in Northeast China.
6Conclusions
By applying observational data of monthly mean land surface
Tmax, Tmin, and Tmean of 118 national stations since 1951,
we investigated spatial and temporal pattern of temperature
change for the periods 1951–2014 and 1998–2014 in
Northeast China. Main conclusions can be drawn as follows:
(1) The increasing trends of Tmax, Tmin, and Tmean were
0.20, 0.42, and 0.34 °C/decade, respectively, over the
period 1951–2014 in Northeast China, with the rate of
Tmin about twice that of Tmax, and the upward trend of
Tmean was clearly higher than the mainland China and
global averages. The long-term warming occurred con-
sistently across the study region, but the largest decrease
of spring mean temperature mainly occurred in central
and southern parts of Northeast China.
(2) In the GWH period, the annual mean temperature con-
sistently exhibited a cooling trend in Northeast China,
and the region-averaged annual trends of Tmax, Tmin,
and Tmean were −0.36, −0.14, and −0.28 °C/decade,
respectively.
(3) Seasonal mean temperature significantly decreased in
winter and spring during the GWH period, but the
warming was still evident in summer and autumn, indi-
cating that the annual mean temperature decrease was
mainly dependent on the remarkable winter and spring
cooling in the study area.
(4) The monthly mean temperature series of Northeast
China showed cooling trends in most months, with the
largest decline occurring in February; however, the
November mean temperature still showed a significant
warming trend.
Fig. 10 A schematic diagram of
the mechanism of winter warming
hiatus phenomenon in Northeast
China during 1998–2014
592 X. Sun et al.
Author's personal copy

(5) Compared with the global and mainland China land sur-
face temperature changes, the hiatus phenomenon in
Northeast China was more evident, and the cooling
trends were most evident in the cold season including
winter and spring.
(6) The Northeast China cooling trend occurs in the time
period of the negative phase Arctic Oscillation, and it is
also closely related to strengthening of the SH and the
EAT. The EAWM obviously becomes stronger over the
GWH period as a result of the abovementioned changes
in the circulation indices.
Acknowledgements This study is financed by the China Natural
Science Foundation (CNSF) (Fund No: 41575003) and the Ministry of
Science and Technology of China (MOST) (Fund No:
GYHY201206012). The authors also thank two anonymous reviewers
for their constructive comments.
References
Alexander LV, Zhang XB, Peterson TC, Caesar J, Gleason B, Klein Tank
AMG, Haylock M, Collins D, Trewin B, Rahimzadeh F (2006)
Global observed changes in daily climate extremes of temperature
and precipitation. J Geophys Res Atmos 111(D5):1042–1063
An WL, Hou SG, Zhang WB, Wu SY, Xu H, Pang HX, Wang YT, Liu YP
(2016) Possible recent warming hiatus on the northwestern Tibetan
Plateau derived from ice core records. Sci Rep 6. doi:10.1038/
srep32813
Balmaseda MA, Trenberth KE, Källén E (2013) Distinctive climate sig-
nals in reanalysis of global ocean heat content. Geophys Res Lett
40(9):1754–1759
Brohan P, Kennedy JJ, Harris I, Tett SFB, Jones PD (2006) Uncertainty
estimates in regional and global observed temperature changes: a
new data set from 1850. J Geophys Res Atmos 111(D12):121–133
Cao L, Zhu Y, Tang G, Yuan F, Yan Z (2016) Climatic warming in China
according to a homogenized data set from 2419 stations. Int J
Climatol 36:4384–4392. doi:10.1002/joc.4639
Carter B (2006) There is a problem with global warming... it stopped in
1998. Telegraph Newspaper (9 April 2006) (http://go.nature.com/
D9CSlh).
Chen S, Chen X, Wei K, Chen W (2014) Vertical tilt structure of EAT and
its interannual variation mechanism in boreal winter. Theor Appl
Climatol 115(3):667–683
Chen X, Tung K (2014) Varying planetary heat sink led to global-
warming slowdown and acceleration. Science 345(6199):897–903
Clark MP, Serreze MC (2000) Effects of variations in East Asian snow
cover on modulating atmospheric circulation over the North Pacific
Ocean. J Clim 13(20):3700–3710
Dong D, Huang G, Qu X, Tao W, Fan G (2015) Temperature trend-
altitude relationship in China during 1963–2012. Theor Appl
Climatol 122(1–2):285–294
Ding Y, Liu Y, Liang S, Ma XQ, Zhang Y, Dong SY, Liang P (2014)
Interdecadal variability of the East Asian winter monsoon and its
possible links to global climate change. J Meteorological Res 28(5):
693–713 (in Chinese)
Golub GH, Reinsch C (1970) Singular value decomposition and least
squares solutions. Numer Math 14(5):403–420
Easterling DR, Wehner MF (2009) Is the climate warming or cooling?
Geophys Res Lett 36(8):262–275. doi:10.1029/2009GL037810
Fang YH, Zhou F, Zhang YF, Zhao ZQ, Shen QY, Wang YS, Wang CX
(2013) The comparison of factors on interannual and interdecadal
variations of the winter temperature in Northeast China. Sci Geogr
Sin 33(7):844–850 (in Chinese)
Franzke CL (2014) Warming trends: nonlinear climate change. Nat Clim
Chang 4(6):423–424
Fyfe JC, Gillett NP (2014) Recent observed and simulated warming. Nat
Clim Chang 4(3):150–151
Fyfe JC, Meehl GA, England MH, Mann ME, Santer BD, Flato
GM, Hawkins E, Gillett NP, Xie SP, Yu K (2016) Making
sense of the early-2000s warming slowdown. Nat Clim
Chang 6(3):224–228
Gong DY, Wang SW, Zhu JH (2001) East Asian winter monsoon and
Arctic oscillation. Geophys Res Lett 28(10):2073–2076
Hansen J, Sato M, Ruedy R, Lo K, Lea DW, Medina-Elizade M (2006)
Global temperature change. Proc Natl Acad Sci 103(39):14288–
14293
Hansen J, Sato M, Kharecha P, Schuckmann KV (2011) Earth’senergy
imbalance and implications. Atmos Chem Phys 11(24):13421–
13449
Huang XM, Guan ZY, Dai ZJ, Mei HX (2013) A further look at the
interannual variations of East Asian trough intensity and their im-
pacts on winter climate of China. Acta Meteorol Sin 71(3):416–428
IPCC (2013) Climate change 2013: the physical science basis. In: Stocker
TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels
A, Xia Y, Bex V, Midgley PM (eds) Contribution of working group I
to the fifth assessment report of the intergovernmental panel on
climate change. Cambridge University Press, Cambridge and New
York, p 1535. doi:10.1017/CBO9781107415324
Jin ZQ, Ge DK, Shi CL, Gao LZ (2002) Several strategies of food crop
production in the Northeast China Plain for adaptation to global
climate change: a modeling study. Acta Agron Sin 28(1):24–31 (in
Chinese)
Jones PD, Osborn TJ, Briffa KR, Folland CK, Horton EB, Alexander LV,
Parker DE, Rayner NA (2001) Adjusting for sampling density in
grid box land and ocean surface temperature time series. J Geophys
Res Atmos 106(D4):3371–3380
Jones PD, Lister DH, Osborn TJ et al (2012) Hemispheric and large-scale
land-surface air temperature variations: an extensive revision and an
update to 2010. J Geophys Res 16(2):206–223. doi:10.1029/
2011JD017139
Kerr RA (2009) What happened to global warming? Scientists say just
wait a bit. Science 326(5949):28–29
Kosaka Y, Xie SP (2013) Recent global-warming hiatus tied to equatorial
Pacific surface cooling. Nature 501(7467):403–407
Karl TR, Arguez A, Huang B, Lawrimore JH, Mcmahon JR, Menne MJ,
Peterson TC, Vose RS, Zhang HM (2015) Possible artifacts of data
biases in the recent global surface warming hiatus. Science
348(6242):1469–1472
Lorenz EN (1956) Statistical forecasting program: empirical orthogonal
functions and statistical weather prediction. Sci Rep 409(2):997–
999
Lathauwer LD, Moor BD, Vandewalle J (2000) A multilinear singular
value decomposition. Siam J Matrix Analysis Applications 21(4):
1253–1278
Lean JL, Rind DH (2009) How will Earth’s surface temperature change in
future decades? Geophys Res Lett 36(15). doi:10.1029/
2009GL038932
Lobell DB, Schlenker W, Costa-Roberts J (2011) Climate trends and
global crop production since 1980. Science 333(6042):616–620
Li JP, Wu ZW (2012) Importance of autumn Arctic sea ice to northern
winter snowfall. Proc Natl Acad Sci 109(109):E1898
Liu Z, Yang X, Chen F, Wang E (2013b) The effects of past climate
change on the northern limits of maize planting in Northeast
China. Clim Chang 117(4):891–902
A remarkable climate warming hiatus 593
Author's personal copy

Li JP, Sun C, Jin FF (2013b) NAO implicated as a predictor of Northern
Hemisphere mean temperature multidecadal variability. Geophys
Res Lett 40(20):5497–5502
Li WJ, Li Y, Chen LJ, Zhao ZG (2013a) Inter-decadal variability of the
relationship between winter temperature in China and its impact
factors. J Appl Meteorological Sci 24(4):385–396 (in Chinese)
Liu S, Sui B, Tu G, Feng XY (2013a) The East Asian winter monsoon
background on the variation of winter air temperature in Northeast
China. J Appl Meteorological Sci 25(1):11–21 (in Chinese)
Li QX, Yang S, Xu WH, Wang XL, Jones P, Parker D, Zhou L, Feng Y,
Gao Y (2015) China experiencing the recent warming hiatus.
Geophys Res Lett 42(3):889–898
Leung YT, Zhou W (2015a) Variation of circulation and East Asian cli-
mate associated with anomalous strength and displacement of the
EAT. Clim Dyn 45(9–10):1–20
Leung YT, Zhou W (2015b) Vertical structure, physical properties, and
energy exchange of the EAT in boreal winter. Clim Dyn 45(5–6):
1635–1656
Leung YT, Cheung HHN, Zhou W (2016) Meridional displacement of
the EAT and its response to the ENSO forcing. Clim Dyn:1–18. doi:
10.1007/s00382-016-3077-8
Meehl GA, Hu A, Arblaster JM, Fasullo J, Trenberth KE (2013)
Externally forced and internally generated decadal climate variabil-
ity associated with the Interdecadal Pacific Oscillation. J Clim
26(18):7298–7310
Qiao SB, Feng GL (2016) Impact of the December North Atlantic
Oscillation on the following February EAT. J Geophys Res Atmos
121. doi:10.1002/2016JD025007
Ren GY, Ding YH, Zhao ZC, Zheng JY, Wu TW, Tang GL, Xu Y(2012)
Recent progress in studies of climate change in China. Adv Atmos
Sci 29(5):958–977
Ren GY, Guo J, Xu MZ, Chu ZY, Zhang LN, Zou XK, Li QX, Liu XN
(2005) Climate changes of mainland china over the past half century.
J Meteorological Res 63(6):942–956 (in Chinese)
Sen PK (1968) Estimates of the regression coefficient based on Kendall’s
tau. J Am Stat Assoc 63(324):1379–1389
Solomon S, Rosenlof KH, Portmann RW, Daniel JS, Davis SM, Sanford
TJ, Plattner G (2010) Contributions of stratospheric water vapor to
decadal changes in the rate of global warming. Science 327(5970):
1219–1223
Solomon S, Daniel JS, Neely RR, Vernier J, Dutton EG, Thomason LW
(2011) The persistently variable Bbackground^stratospheric aerosol
layer and global climate change. Science 333(6044):866–870
Sun C, Li JP (2012) Analysis of anomalously low surface air temperature
in the Northern Hemisphere during 2009/2010 winter. Clim Environ
Res 17(3):259–273 (in Chinese)
Santer BD, Bonfils C, Painter JF, Zelinka MD, Mears C, Solomon S,
Schmidt GA, Fyfe JC, Cole JN, Nazarenko L (2014) Volcanic con-
tribution to decadal changes in tropospheric temperature. Nat Geosci
7(3):185–189
Seneviratne SI, Donat MG, Mueller B, Alexander LV (2014) No pause in
the increase of hot temperature extremes. Nat Clim Chang 4(3):161–
163
Sun C, Li JP, Jin FF (2015) A delayed oscillator model for the quasi-
periodic multidecadal variability of the NAO. Clim Dyn 45:2083–
2099. doi:10.1007/s00382-014-2459-z
Sun Y, Zhang XB, Ren GY, Zwiers FW, Hu T (2016) Contribution of
urbanization to warming in China. Nat Clim Chang. doi:10.1038/
nclimate2956
Sun XB, Ren GY, Xu WH, Li QX, Ren YY (2017) Global land-surface
air temperature change based on the new CMA GLSAT dataset. Sci
Bull 62(4):236–238
Thompson DWJ, Wallace JM (1998) The Arctic oscillation signature in
the wintertime geopotential height and temperature fields. Geophys
Res Lett 25(9):1297–1300
Tang HY, Zhai PM, Wang ZY (2005) On change in mean maximum
temperature minimum and diurnal range in china during 1951–
2002. Clim Environ Res 10(4):728–735 (in Chinese)
Tang GL, Luo Y, Huang JB, Wen XY, Zhu YN, Zhao ZC, Wang SW
(2012) Continuation of the global warming. Adv Clim Chang Res
8(4):235–242 (in Chinese)
Tollefson J (2014) Climate change: the case of the missing heat. Nature
505(7483):276–278
Trenberth KE, Fasullo JT, Balmaseda MA (2014a) Earth’s energy imbal-
ance. J Clim 27(9):3129–3144
Trenberth KE, Fasullo JT, Branstator G, Phillips AS (2014b) Seasonal
aspects of the recent pause in surface warming. Nat Clim Chang
4(10):911–916
Wu BY, Wang J (2002) Winter arctic oscillation, siberian high an East
Asian winter monsoon. Geophys Res Lett 29:1897. doi:10.1029/
2002GL015373
Wang H, He S (2012) Weakening relationship between East Asian winter
monsoon and ENSO after mid-1970s. Chin Sci Bull 57:3535–3540
Yan L, Liu X (2014) Has climatic warming over the Tibetan Plateau
paused or continued in recent years? Journal of Earth. Ocean
Atmos Sci 1(1):13–28
Zhu YF, Tan GR, Wang YG (2008) An index of East Asian winter mon-
soon applied to description the Chinese mainland winter tempera-
ture changes. Acta Meteor Sin 66(5):781–788 (in Chinese)
Zhou Q, Wen C, Wen Z (2013) Solar cycle modulation of the ENSO
impact on the winter climate of East Asia. J Geophys Res Atmos
118(118):5111–5119
Zhao CY (2016) Characteristics and long-term trend of plant phonology
in Northeast China. A Ph.D. thesis of the Nanjing University of
Information Science &Technology, Nanjing, China.
Zhao S, Li JP, Sun C (2016) Decadal variability in the occurrence of
wintertime haze in central eastern China tied to the Pacific
Decadal Oscillation. Sci Rep 6:27424. doi:10.1038/srep27424
594 X. Sun et al.
Author's personal copy