Clim Dyn (2012) 39:2487–2496 DOI 10.1007/s00382-012-1354-8

How does coldwave frequency in china respond to a warming climate? Tingting Ma • Zhiwei Wu • Zhihong Jiang

Received: 22 August 2011 / Accepted: 21 March 2012 / Published online: 5 April 2012 Ó Springer-Verlag 2012

Abstract Under the background of a warming climate, regional climate responses may be different from place to place. How cold extreme events in China respond is still an open question. This study investigates responses of coldwave frequency (CWF) in China from observation and modeling perspectives. Observational evidences show that CWF significantly reduces across China during the warm period (1978–2009) in comparison with that during the cold period (1957–1977), concurrent with extreme value centers located in northern China during 1957–1977 and southern China during 1978–2009. The empirical orthogonal function (EOF) leading mode of CWF in the cold period is also dominant by an extreme value center prevailing over northern China, while the center exhibits a southward shift in the warm period. A seven-member multi-model ensemble (MME) from coupled model intercomparison project#3 (CMIP3) shows that southern China tends to experience more coldwaves than northern China in the twenty first century (2045–2064 and 2080–2099) under the global warming A1B forcing (with atmospheric CO2 concentration of 720 ppm). This feature can also be seen in the leading EOF mode of MME. These results indicate that the primary response of CWF to a warming climate may be the southward shift of the maximum loading center. The

T. Ma  Z. Jiang Key Laboratory of Meteorological Disaster of Ministry of Education, Nanjing University of Information Science and Technology, Nanjing, China Z. Wu (&) Meteorological Research Division, Environment Canada, Dorval, QC H9P 1J3, Canada e-mail: [email protected]

enhanced western Pacific Subtropical High and weakened Siberian High during 1978–2009 may result in anomalous southerlies which bring warm and wet air to southern China. Meanwhile cold and dry air is transported from the north via a ‘‘northwest pathway’’ to southern China. Under the joint action of these two air masses, coldwaves may easily generate in southern China as observed in recent extreme cold events in this region. Keywords Coldwave frequency  Global warming  Climate change

1 Introduction The earth’s climate is experiencing significant changes characterized by global warming. The global mean temperature has increased 0.74 °C (±0.18 °C) in the last 100 years (1906–2005). According to the Fourth Assessment Report coordinated by the Intergovernmental Panel on Climate Change (IPCC-AR4), the linear warming trend over the 50 years from 1956 to 2005 is nearly twice that for the 100 years from 1906 to 2005 (Solomon et al. 2007). Under the background of a warming period, responses of regional climate, including extreme climate events, may be different from place to place. Quite a few studies have focused on the responses of extreme warm events (e.g., Karl et al. 1996; Karl and Knight 1997; Karl and Easterling 1999; Meehl and Tebaldi 2004; Scha¨r et al. 2004; Clark et al. 2006; Garcı´a-Herrera et al. 2010). For instance, during the 1995 Chicago heatwave, the temperature soared to record highs, which was the second warmest July temperature since records began at Chicago Midway International Airport in 1928 (Karl and Knight 1997). Scha¨r et al. (2004) and Clark et al. (2006) suggested that European

123

2488

summer heatwaves will become more frequent and severe during this century. Extreme cold events have received increasing attention during the past few years due to their frequent occurrences (e.g., Wen et al. 2009; Wu et al. 2010; Cattiaux et al. 2010; Chen et al. 2010; Jiang et al. 2011). In 2008 (January– February), East Asia (particularly southern China) experienced the most severe and long-persisting snowstorm in the past 100 years (e.g., Wen et al. 2009; Hong and Li 2009; Wu et al. 2010). In 2009/2010 winter, several Northern Hemisphere countries were punctuated by series of extreme cold events. In Europe, several severe cold spells were observed over Northern and Western Europe (Cattiaux et al. 2010). In North America, several regions recorded their snowiest winter ever (Seager et al. 2010), while a few cities witnessed record-breaking cold daily temperature (Wang et al. 2010b; Lin and Wu 2011a, b). Kodra et al. (2011) found that in many places, cold extremes may not be less severe or shorter in Twenty first century than they have been in the recent past on average. How the cold extremes in China respond to a warming climate is still an open question. After the record-breaking snowstorm occurred in southern China in early 2008, most studies regarded this event as a once-in-a-century case and therefore focused on its unique features (e.g., Tao and Wei 2008; Ding et al. 2008; Bao et al. 2010). However, Wu et al. (2010) and Zhang et al. (2011) pointed out that although the catastrophic consequences of the 2007/2008 snowstorms are unprecedented, several similar cases do occur in the past 50 years. In January 2010, southern China experienced extreme cold weathers once again (http:// ncc.cma.gov.cn/upload/upload2/jdjc/qhsj_m110100.bmp). It seems that extreme cold events are more likely to occur in southern China in recent decades. This is repugnant to our common sense that northern China should suffer more coldwavs. Under a warming background, what happens to cold extremes in China during the past decades? Is the southward shift of the coldwaves the response to a warming climate? This study attempts to answer the above questions. This paper is structured as follows. Section 2 describes the datasets, models and the methodology used in this study. Section 3 investigates the observed coldwave frequency (CWF) variations in the warming period. To further verify the conclusion derived from the observations, the response of CWF is investigated with the A1B stabilization experiments (with an atmospheric CO2 concentration of 720 ppm) for twenty first century (2045–2064, 2080–2099) conducted by seven coupled models from the coupled model intercomparison project#3 (CMIP3, Sect 4). Section 5 discusses the possible physical mechanisms on the response of CWF. The last section summarizes major findings and discusses some outstanding issues.

123

T. Ma et al.

2 Data, model and methodology The major datasets used include (1) daily surface air temperature data at 605 gauge stations across China from China meteorological administration (see Fig. 1); (2) normalized global mean surface air temperature data from the University of East Anglia Climate Research Unit (CRU) (Jones et al. 1999); (3) monthly geopotential height data, sea level pressure (SLP) data, and wind data, gridded at 2.5° 9 2.5° resolution, taken from the European centre for mediumrange weather forecasts (ECMWF) 40-year reanalysis dataset (ERA-40; Uppala et al. 2005) and the National Oceanic and Atmospheric Administration (NOAA) national centers for environmental predictions reanalysis data (NCEP-DOE reanalysis 2) (Kanamitsu et al. 2002). Global warming A1B stabilization experiments (with an atmospheric CO2 concentration of 720 ppm) for twenty first century (2045–2064, 2080–2099) used in this study are conducted by 7 state-of-the-art coupled models that participated in IPCC-AR4. These models include (1) Max Planck Institute (MPI), ‘‘ECHAM5’’/Max Planck Institute Ocean Model (MPI-OM), (2) Meteorological Research Institute Coupled General Circulation Model, version 2.3.2 (MRICGCM2.3.2), (3) Canadian Centre for Climate Modelling and Analysis Coupled General Circulation Model, version 3.1 [CCCma-CGCM3.1(T63)], (4) Centre National de Recherches Me0 te0 orologiques Coupled Global Climate Model, version 3 (CNRM-CM3), (5) Commonwealth Scientific and Industrial Research Organisation, Mark version 3.0 (CSIRO-Mk3.0), (6) Center for Climate System Research (The University of Tokyo), National Institute for Environmental Studies, and Frontier Research Center for Global Change (JAMSTEC), hires version 3.2 (MIROC3.2hires), (7) Max Planck Institute for Meteorology, ECHAM4. The credibility of simulated cold extremes is evaluated

Fig. 1 Distribution of 605 surface air temperature gauge stations across China

How does coldwave frequency

2489

3.1 Evolution of the CWF Climatology

Fig. 2 November–February (NDJF) normalized global mean surface air temperature (color bars) from the CRU dataset (Jones et al. 1999) for 1957–2010. The black curve denotes decadal variations obtained using a nine-point Gaussian-type low-pass filter

through both bias scores relative to reanalysis data in the past and multi-model agreement in the future (Kodra et al. 2011). In this paper, winter refers to November–February (NDJF). The observational data used in this study cover the period from November 1957 to February 2010, and the modeling data from 2045 to 2065, and 2080 to 2100. We employed ERA-40 reanalysis data for the period 1957–2002 and extended the data from 2003 to 2010 by using NCEP-DOE reanalysis 2 data. To maintain temporal homogeneity, the 2003–2010 NCEP-DOE reanalysis 2 data were adjusted by removing the climatological difference between the ERA-40 and NCEP-DOE reanalysis 2 datasets (Li et al. 2010; Wang et al. 2010a). According to the definition of heatwaves by Fischer and Scha¨r (2010), we define a coldwave as a spell of at least six consecutive days with minimum temperatures less than the local 10th percentile of the control period (1961–1990). To account for the seasonal cycle, the 10th percentile is calculated for each calendar day, and at each grid point using a centered 15-day-long time window. The CWF refers to the average frequency of days meeting the coldwave criterion. Considering the global mean temperature transition from a cold stage to a warm stage (Fig. 2), the periods 1957–1977 and 1978–2010 are treated as cold and warm epochs for the following analysis, respectively.

Figure 3 presents the climatology of CWF for the cold and warm epochs, respectively. The CWF significantly reduces across China during 1978–2009 (Fig. 3b) in comparison with that during 1957–1977 (Fig. 3a). The annual average value of CWF in the cold period is around 12 days, while in the warm period it decreases to around 6 days. Another prominent feature is that the spatial distributions of CWF in the two epochs are different. The CWF high-value center lies in northern China during the cold period, with lowvalue areas located in the south, whereas during the warm epoch, the high values are centered over southern China and a few places in the north. The change in climatology from the cold period through the warm period indicates southern China is suffering more cold extremes than northern China in a warming climate. For further comparison, we calculate the difference and the ratio of CWF between the two epochs. Figure 4 displays the composite difference of CWF between the cold and warm periods (warm minus cold). The CWF variability basically exhibits the same sign, with negative extreme value centers located over northern China. It implies that CWF in the warm period diminishes across China compared with that in the cold period and the reduction in the north is much larger than that in the south. The negative

(a)

(b)

3 Observed CWF variations in a warming background How does CWF in China evolve during the past decades? To figure out this question, we examine the CWF climatology and its empirical orthogonal function (EOF) dominant modes for the two periods 1957–1977 and 1978–2010.

Fig. 3 NDJF coldwave frequency (CWF, units: days) across China in a the cold period (1957–1977) and b the warm period (1978–2009)

123

2490

Fig. 4 NDJF CWF composite difference between the cold and the warm periods (the latter minus the former). Color shadings denote negative CWF anomalies exceeding the 90 % confidence level

Fig. 5 Ratio of CWF between the cold and warm periods (warm/cold epochs)

CWF anomalies in some north regions and the east coast exceed the 90 % confidence level. From the ratio of CWF between the two periods (warm/ cold epoch) shown in Fig. 5, we may further quantify the changes of CWF all over the country. The ratio in northern and eastern China is around 0.4, which means in these regions, amount of CWF during 1978–2009 is only 40 % of that during 1957–1977. The ratio in southern China is greater than that in northern China, which means that CWF reduces much more in northern China than in southern China. All of these findings are consistent with the above composite analysis results. 3.2 Evolution of CWF leading modes To derive the dominant modes of CWF, we perform EOF analysis of CWF in the cold and warm periods, respectively. The two leading modes (hereafter EOF1 and EOF2) are shown in Fig. 6.

123

T. Ma et al.

In the cold period, the two leading modes account for 40.2 and 12.3 % of the total variance, respectively (Fig. 6a, c). According to the rule given by North et al. (1982), the two modes are statistically distinguished from each other. The spatial pattern of the EOF1 mode shows a mono-sign pattern, with the extreme value center prevailing over northern China. In fact, the EOF1 mode corresponds to the middle and eastern trajectory of cold air masses affecting China (e.g. Li 1955; Tao 1957; Wu et al. 2009; Wang et al. 2010a), which can influence the majority of China, yet with a decreasing trend in the past 50 years. The EOF2 mode primarily shows a tripole pattern with two maximum loadings located over northeastern and southern China. In the warm period, the EOF1 mode accounts for 23.9 % of the total variance, and the EOF2 mode 16.5 % (Fig. 6b, d). The leading mode also shows a mono-sign pattern, except a negative anomaly in small regions of northeastern China. It is interesting to notice that the extreme value centers of EOF1 are located in southern China, which is consistent with the climatology results shown in Fig. 3b. The EOF1 mode represents the western cold air mass trajectory (e.g. Li 1955; Tao 1957; Wu et al. 2009; Wang et al. 2010a), which basically affects southern China. The EOF2 mode features a dipole pattern with negative CWF anomalies prevailing over the southeast coastal area. From the above analysis, we can conclude that CWF significantly reduces across China during the warm period. Both the CWF climatology and the leading modes are characterized by extreme value centers located in northern China during the cold epoch (1957–1977) and southern China during the warm epoch (1978–2009), which indicates that southern China is experiencing more coldwaves than northern China in a warming climate.

4 Future change of CWF From the observational perspective, we find that the extreme value center of CWF shifts to southern China from the cold period to the warm one. We suppose this southward shift is the response of CWF to a warming climate. We may then speculate that, with the increasing of temperature, the maximum loading center of CWF may accentuate in southern China in the future. To investigate this issue, we use simulations of seven state-of-the-art coupled climate models from CMIP3. Recent attribution studies (Pierce et al. 2009; Santer et al. 2009) in the context of regional climate suggest that the multi-model ensemble (MME) works better than any individual model does. We use the MME in our study. To evaluate models cold extremes, Fig. 7 provides a comparison between observed and simulated CWF in 1961–1990. Observed CWF decreases from north to south.

How does coldwave frequency Fig. 6 Spatial pattern of CWF leading mode in a the cold period and b the warm period. c and d same as in (a) and (b) but for the second mode. Color shadings have the same units that are used in Fig. 3

Fig. 7 a NDJF Climatological CWF across China in the observation for the 1961–1990 period. b Same as in (a) but for multi-model ensemble (MME) of seven coupled model intercomparison project#3 (CMIP3) models

2491

(a)

(b)

(c)

(d)

(a)

The annual average value of CWF is around 10 days. The distribution of simulated CWF is similar to the observational counterpart, except for the eastern side of Tibetan Plateau. The mean value of CWF is also around 10 days. We can conclude that these models have a reasonable skill in simulating CWF. To verify whether CWF extreme value center shifts to southern China in the twenty first century, we calculate the changes of CWF in 2045–2064 and 2080–2099 from the control period 1961–1990. Figure 8 presents the distribution of CWF across China in 2045–2064 and 2080–2099, respectively. It is interesting to notice that the two periods display a similar distribution, with the extreme value center located in southern China. A small difference can be however observed: the maximum loading center is located in southwest China in 2045–2064, but southeast China in 2080–2099. Figure 8 is in favor of our previous assumption that the maximum loading center

(b)

of CWF may accentuate in southern China with the anthropogenic global warming. We also perform EOF analysis of simulated CWF in the two periods, respectively. Results are shown in Fig. 9. The EOF1 mode explains 48.8 % of the total variance in 2045–2064, and 39.4 % in 2080–2099. The leading modes in the two periods show both a mono-sign pattern, with the largest positive anomaly around southern China, which further verifies our assumption. The EOF1 mode in 2045–2065 has another maximum loading center located over northern China. It may be caused by the uncertainty in the model outputs. These model projections provide further evidences that the southward shift of the CWF extreme value center might be the response of the CWF to a warming climate. Our results suggest that cold extremes are more likely to occur in southern China in a warming period.

123

2492

T. Ma et al.

Fig. 8 Changes of NDJF climatological CWF of a 2045–2064 versus b 2080–2099

(a)

(a)

(b)

Fig. 9 Spatial pattern of the first EOF mode for a 2045–2064, b same as in (a) but for 2080–2099

5 Possible physical mechanisms As noted above the response of CWF to a warming climate might be the southward shift of the extreme value center, the following will focus on why CWF have such response? In this section, we discuss the possible physical processes causing this change in the two different periods. Firstly, we present the planetary-scale atmospheric circulation in the two periods. Figure 10 shows the winter atmospheric circulation and its difference between the cold

123

(b)

and warm epochs (warm minus cold). Near the surface (Fig. 10g, h), Eurasian continent is mainly controlled by Siberian High (SH), and North Pacific is covered by Aleutian Low (AL) in both the cold and warm periods. It is clear that a large area of significant negative SLP anomalies occupies the entire northern Eurasian continent (Fig. 10i), with a major trough extending southward along the eastern flank of the Tibetan Plateau (TP), indicating a weakened SH in the warm period. A negative SLP anomaly center although weak, can be seen over the North Pacific. Several areas of salient positive SLP anomalies mainly lie in the low latitudes. At the middle and upper troposphere (Fig. 10c, f), two anomalous high pressure centers are located over Lake Baikal and southern China, exhibiting the baroclinic structure of SH. The anomalous high over Lake Baikal implies an eastward shift of the East Asian (EA) major trough. A non-significant low pressure anomaly is seen along the western of Ural Mountain and a low pressure center occupies the North Pacific. Notable positive high anomalies primarily control the region south of 20°N. The three-dimensional circulation pattern is corresponding to a weakened SH and a strengthened western Pacific Subtropical High (WPSH) in the warm period. The SH is a strong cold-core high pressure. Associated with it, the majority of EA continent are occupied by northerly surface winds (Ding 1994), which bring cold and dry air mass from the north. In general, cold and dry air from the north is responsible to cold surface temperature anomalies (Zhang et al. 2011). Recent extreme cold events in southern China are often accompanied with wet weather. The weakened SH may cause anomalous southerlies, which bring warm and moisture air from south. The enhanced WPSH may also result in anomalous southerlies, which is necessary for the generation of the wet and cold weathers. Under the joint action of the cold and dry air from the north and the warm and moisture air from the south, the cold and wet weathers may easily form. To assess the possible physical processes responsible for the leading mode change, we present the anomalous circulation regressed on the PC1 during the two periods,

How does coldwave frequency (a)

(d)

(g)

2493 (b)

(c)

(e)

(f)

(h)

(i)

Fig. 10 Atmospheric general circulations in winter depicted at 200 hPa (top panels, geopotential height in contour in 10 m and winds in m/s), 500 hPa (middle panels, geopotential height in contour in m and winds in m/s) and near the surface (bottom panels, sea-level pressure in contour in hPa and winds at 925-hPa in m/s) respectively.

Left panels are from the cold period and middle ones the warm period. Right panels show the difference (warm minus cold) of the geopotential height (or SLP). Blue (yellow) shadings denote negative (positive) anomalies exceeding the 90 % significance level, based on the Student-t test

respectively. During the cold epoch (1957–1977), a gigantic anomalous high SLP occupies the entire Eurasian continent, with a major ridge extending southward along the eastern flank of TP to South China Sea (Fig. 11g). A strong low SLP anomaly is seen over North Pacific. The northerly winds near the surface prevail in EA, which advect cold air southward from north. At the middle and upper troposphere (Fig. 11a, d), the salient positive pressure anomaly lies in the region north of 40°N, and the anomalies low control the continent south of 40°N and North Pacific. The large meridional gradient between the anomalous high and low implies strong anomalous northerlies from central Siberia to EA. During the warm epoch (1978–2009), a low pressure anomaly covers almost the whole Eurasian continent and North Pacific (Fig. 11b, e). A high pressure anomaly is found only in a small area nearby 60°N in the middle and upper troposphere. Near the surface (Fig. 11h), large areas of positive SLP anomalies are located over the continent region south of 60°N and the

ocean in the low latitudes, rather than the whole Eurasian continent in the cold period. Areas in higher latitude are occupied by negative SLP anomaly. Such a feature resembles the structure identified as the south mode in Wang et al. (2010a). This pattern reflects a route of cold-air intrusion along the eastern flank of TP via a ‘‘northwest pathway’’ to Southeast Asia. As mentioned above, warm and moisture air from the south is prevailing in southern China. Under the cooperating action of the cold and dry air and the warm and moisture air, the cold and wet weathers may easily occur in southern China. It could be clearly seen in the difference of anomalous circulation pattern between the cold and warm periods in Fig. 11c, f, i. A large anomalous low SLP covers the entire Eurasian continent, indicating a weakened SH in the warm periods. A weakened AL is found over the North Pacific. At the upper levels, low pressure anomaly controls the area north of 50°N. China and North Pacific are occupied by high pressure anomalies. This pattern implies a weakened East

123

2494

T. Ma et al.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

Fig. 11 a 200 hPa geopotential height (contours in unit of hPa) and winds (vectors in unit of m/s) anomalies regressed on the PC1 in the cold period, b 200 hPa geopotential height and winds anomalies regressed on the PC1 in the warm period and c the composite

difference of geopotential height anomalies between the two epochs (warm minus cold period). d–f Same as in (a), (b) and (c) but for 500 hPa geopotential height and winds. (g–i) Same as in (a), (b) and (c) but for SLP and 925 hPa winds

Asian winter monsoon (EAWM) associated with the dominant mode of CWF in the warm period. We also present the anomalous circulation regressed on PC2 in Fig. 12. In the cold period, a gigantic anomalous high SLP occupies most of the EA continent, with a major ridge extending southward along the Ural Mountain. Another high SLP anomaly is centered over the subtropical North Pacific. A negative SLP anomaly center is seen over the northern North Pacific. A huge elongated anticyclonic anomaly covers all of high-latitude Russia and the northern North Pacific at the middle and upper troposphere. This circulation pattern reflects an enhanced EAWM. The large zonal gradient between the anomalous EA continent high and the North Pacific low implies strong anomalous northerlies which affect northern China. The anomalous circulation pattern in the warm epoch is similar to that in the cold epoch. But the anomalous pressure in high latitude

in the warm period is much higher than that in the cold period, as we can see from Fig. 12c, f, i. As observed in recent extreme cold events in southern China, coldwaves are often accompanied with cold and wet weathers. Cold and dry air from the north is generally responsible to cold surface temperature anomalies. In the warm period, the weakened SH and the enhanced WPSH may result in anomalous southerlies. The anomalous southerlies may bring warm and moisture air to southern China, which is necessary for the generation of wet and cold weathers. Meanwhile cold and dry air is transported along the eastern flank of TP via a ‘‘northwest pathway’’ to southern China. Under the joint action of the cold and dry air from the north and the warm and moisture air from the south, coldwaves accompanied with wet weather may easily generate in southern China as observed in recent extreme cold events in this region.

123

How does coldwave frequency

2495

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

Fig. 12 As in Fig. 11, but for regressed on PC2

6 Conclusion and discussion The warming climate brings about many pronounced changes in the global climate (e.g., Lau et al. 2008; Li et al. 2010), what kind of influences it may exert on winter climate in China is still an outstanding issue. Using observed data and model simulations, we attempt to unveil the responses of CWF in China to a warming climate. The CWF significantly decreases across the whole country during the warm period in comparison with that during the cold period. Observational evidences and model projections indicate that the primary response of CWF may be the southward shift of the extreme value center. On account of such southward shift, southern China is likely to experience more cold extremes in a warming climate. The possible physical processes are also discussed. The enhanced WPSH and weakened SH during 1978–2009 may result in anomalous southerlies in southern China, which is necessary for the generation of wet and cold weathers. Meanwhile cold and dry air is transported from the north along the eastern flank of TP via a ‘‘northwest pathway’’ to southern China. Under the joint action of the cold and dry

air and the warm and moisture air, coldwaves accompanied with wet weather may easily generate in southern China as observed in recent extreme cold events in this region. Because what we emphasize in this study is to point out the response of CWF to a warming climate, there is a lack of the possible physical processes. Many detailed physical processes are not fully understood. For example, what leads to the changes in SH and WPSH? How can the seasonal EAWM change affect coldwaves in China? Li et al. (2010) investigated the responses of the East Asian Summer Monsoon (EASM) from observations, theoretical and modeling perspectives and pointed out that the principal response of the EASM is the southward shift in the major rain belts and the associated circulations. Whether their theory can also explain the southward shift of CWF extreme value center? These issues may be investigated in future. Some of them will be discussed in other papers. Acknowledgments We thank Dr. Laurent Li from Laboratoire de Me´te´orologie Dynamique, IPSL/CNRS/UPMC, Paris, France and the two anonymous reviewers for helpful and valuable comments. The authors are supported by the National Basic Research Program ‘‘973’’ (Grant No. 2010CB950401 and 2010CB428505), the Special

123

2496 Research Program for Public Welfare (Meteorology) of China under Grant No. GYHY200906016 and a Project Funded by the Priority Academic Program Development (PAPD) of Jiangsu Higher Education Institutions.

References Bao Q, Yang J, Liu YM, Wu GX, Wang B (2010) Roles of anomalous Tibetan Plateau warming on the severe 2008 winter storm in central-southern China. Mon Weather Rev. doi:10.1175/2009 MWR2950.1 Cattiaux J, Vautard R, Cassou C, Yiou P, Masson-Delmotte V, Codron F (2010) winter 2010 in Europe: a cold extreme in a warming climate. Geophys Res Lett 37:L20704. doi:10.1029/ 2010GL044613 Chen WL, Jiang ZH, Li L, Yiou P (2010) Simulation of regional climate change under the IPCC A2 scenario in southeast China. Clim Dyn. doi:10.1007/s00382-010-0910-3 Clark RT, Brown SJ, Murphy JM (2006) Modeling Northern Hemisphere summer heat extreme changes and their uncertainties using a physics ensemble of climate sensitivity experiments. J Clim 19:4418–4435 Ding YH (1994) Monsoons over China. Kluwer Academic Publisher, Dordrecht Ding YH, Wang ZY, Song YF, Zhang J (2008) Causes of the unprecedented freezing disaster in January 2008 and its possible association with the global warming. Acta Meteorol Sinica 665:809–825 Fischer EM, Scha¨r C (2010) Consistent geographical patterns of changes in high-impact European heatwaves. Nat Geosci 3:398–403 Garcı´a-Herrera R, Dı´az J, Trigo RM, Luterbacher J, Fischer EM (2010) A review of the European summer heatwave of 2003. Crit Rev Environ Sci Technol 40:267–306 Hong CC, Li T (2009) The extreme cold anomaly over southeast Asia in February 2008: roles of ISO and ENSO. J Clim 22:3786–3801 Jiang ZH, Song J, Li L, Chen WL, Wang ZF, Wang J (2011) Extreme climate events in China: IPCC-AR4 model evaluation and projection. Clim Change. doi:10.1007/s10584-011-0090-0 Jones PD, New M, Parker DE, Martin S, Rigor IG (1999) Surface air temperature and its variations over the last 150 years. Rev Geophys 37:173–199 Kanamitsu M, Ebisuzaki W, Woollen J, Yang S-K, Hnilo JJ, Fiorino M, Potter GL (2002) NCEP-DOE AMIP-2 Reanalysis (R-2). Bull Atmos Met Soc 83:1631–1643 Karl TR, Easterling DR (1999) Climate extremes: selected review and future research directions. Clim Change 42:309–325 Karl TR, Knight RW (1997) The 1995 Chicago heat wave: how likely is a recurrence? Bull Am Meteorol Soc 78:1107–1119 Karl TR, Knight RW, Easterling DR, Quayle RG (1996) Indices of climate change for the United States. Bull Am Meteorol Soc 77:279–291 Kodra E, Steinhaeuser K, Ganguly AR (2011) Persisting cold extremes under twenty first-century warming scenarios. Geophys Res Lett 38:L08705. doi:10.1029/2011GL047103 Lau NC, Leetmaa A, Nath MJ (2008) Interactions between the responses of North American climate to El Nin˜ o-La Nin˜ a and to the secular warming trend in the Indian-Western Pacific Oceans. J Climate 21:476–494 Li XZ (1955) A study of cold waves in East Asia. In: Li XZ (ed) Offprints of scientific works in modern China-meteorology (1919–1949). Science Press, Beijing, pp 35–117 (in Chinese)

123

T. Ma et al. Li JP, Wu ZW, Jiang ZH, He JH (2010) Can global warming strengthen the East Asian summer monsoon? J Clim 23:6696– 6705 Lin H, Wu ZW (2011a) Contribution of the autumn Tibetan Plateau snow cover to seasonal prediction of North American winter temperature. J Clim 24:2801–2813 Lin H, Wu ZW (2011b) Contribution of Tibetan Plateau snow cover to the extreme winter condition of 2009–2010, Atmos-Ocean, revised Meehl GA, Tebaldi C (2004) More intense, more frequent, and longer lasting heat waves in the twenty first century. Science 305:994–997 North GR, Bell TL, Cahalan RF, Moeng FJ (1982) Sampling errors in the estimation of empirical orthogonal functions. Mon Weather Rev 110:699–706 Pierce DW, Barnett TP, Santer BD, Gleckler PJ (2009) Selecting global climate models for regional climate change studies. Proc Natl Acad Sci USA 106(21):8441–8446 Santer B et al. (2009) Incorporating model quality information in climate change detection and attribution studies. Proc Natl Acad Sci USA 106(35):14,778–14,783. doi:10.1073/pnas.0901736106 Scha¨r C et al (2004) The role of increasing temperature variability in European summer heatwaves. Nature 427:332–336 Seager R, Kushnir Y, Nakamura J, Ting M, Naik N (2010) Northern Hemisphere winter snow anomalies: ENSO, NAO and the winter of 2009/10. Geophys Res Lett 37:L14703. doi:10.1029/2010 GL043830 Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M, Miller HL (eds) (2007) Climate change 2007: the synthesis report. Contribution of working group I to the fourth assessment report of the intergovernmental panel on climate change. Cambridge University Press, Cambridge, UK and New York, NY, USA, 996 p Tao SY (1957) A study of activities of cold airs in East Asian winter, handbook of short-term forecast (in Chinese), china meteorological administration. Meteorology Press, Beijing, pp 60–92 Tao SY, Wei J (2008) Severe snow and freezing rain in January 2008 in the southern China. Climatic Environ Res 13(4):337–350 (in Chinese) Uppala S, Kallberg PW, Simmons AJ et al (2005) The ERA-40 re-analysis. Quart J Roy Meteor Soc 131:2961–3012. doi: 10.1256/qj.04.176 Wang B, Wu ZW, Chang CP, Liu J, Li JP, Zhou TJ (2010a) Another look at climate variations of the East Asian winter monsoon: Northern and Southern modes. J Clim 23:1495–1512 Wang C, Liu H, Lee S (2010b) The record-breaking cold temperatures during the winter of 2009/2010 in the Northern Hemisphere. Atmos Sci Lett. doi:10.1002/asl.278 Wen M, Yang S, Kumar A, Zhang P (2009) An analysis of the largescale climate anomalies associated with the snowstorms affecting China in January 2008. Mon Weather Rev 137:1111–1131 Wu ZW, Li JP, Wang B, Liu XH (2009) Can the Southern Hemisphere annular mode affect China winter monsoon? J Geophys Res 114:D11107. doi:10.1029/2008JD011501 Wu ZW, Li JP, Jiang ZH, He JH (2010) Predictable climate dynamics of abnormal East Asian winter monsoon: once-in-a-century snowstorms in 2007/2008 winter. Clim Dyn. doi:10.1007/s00382010-0938-4 Zhang HQ, Qin J, Li Y (2011) Climatic background of cold and wet winter in southern China: part 1 observational analysis. Clim Dyn. doi:10.1007/s00382-011-1022-4