Abstract

The recent increase in the frequency of winter cold extremes has received particular attention in light of the climate's warming. Knowledge about changes in the frequency of winter cold extremes requires long-term climate data over large spatial scale. In this study, a temperature-sensitive tree-ring network consisting of 31 sampling sites collected from seven provinces in subtropical China was used to investigate the characteristics of cold-season temperature extremes during the past two centuries. The results show that the percentage of trees in a year that experienced an abnormal decrease in radial growth relative to the previous year can serve as an indicator of interannual change in January–March temperature in subtropical China. The frequency of extreme interannual decreases in cold-season temperature has increased since the 1930s. The change in cold-season temperature was significantly correlated with the intensity of the Siberian high, yet the correlation was much weaker in the period preceding the 1930s. The findings provide evidence of a frequency change in the occurrence of interannual cold-season temperature extremes in the past two centuries for subtropical China. Particularly, the pattern in the variation of cold-season temperature suggests a change in climate systems around the 1930s.

1. Introduction

Extreme winter coldness is among the most destructive disasters that severely affects people's lives and economic development in many parts of the world. Such extreme coldness with heavy snowfall occurred in Europe and North America in the winters of 2009, 2010, and 2011 (Seager et al. 2010; Ratnam et al. 2012; Moore and Renfrew 2012; Li 2012; Cameron 2012). In subtropical China, which is a region with dense population and rapid economic development, an extreme coldness hit the region in January–February 2008, causing great damage to agriculture, transportation, and power infrastructure (Ding et al. 2008; Zhao et al. 2008). Repeatedly, extreme coldness impacted the region in January 2011 and January–February 2012. The occurrence of these consecutive coldness events in recent years has received great attention from climate researchers as well as the public. Two questions arise inevitably: 1) Is the occurrence of cold-season temperature extremes increasing in frequency? 2) How is the extreme winter coldness linked to global climate systems especially in the course of global warming?

The extreme coldness events in subtropical China were studied using observed climate records. Zhang et al. (2009) showed that extremely cold Januaries occurred frequently in the 1960s in southern China over the period 1951–2008. Liu and Gao (2001) found that spring cold extremes in southern China were more frequent in the 1980s than in the 1990s. These studies indicated that the frequency of extreme coldness events was uneven through time and that different synoptic systems could induce extreme coldness in the region. The relationships between winter climate in East Asia and several climate systems (e.g., the Siberian high, the Arctic Oscillation, and the East Asian winter monsoon) have been reported in previous studies (Gong et al. 2004; Wen and Chen 2006; Li et al. 2007; Huang et al. 2007; Zhu 2008; Wang et al. 2010). However, these results were restricted to the interval of instrumental records, which is too short to detect the full spectrum of the frequency characteristics of extreme climate events and its possible linkage to climate systems. To extend climate information beyond the instrumental record, tree rings were used to reconstruct past climate for the region (Shi et al. 2010; Duan et al. 2012). However, these previous tree-ring studies were limited by weak climate signals and the small spatial coverage of the tree-ring samples. To date, large-scale and long-term characteristics of cold-season temperature variations still remain poorly understood for subtropical China.

In this study, we developed a two-century-long cold-season temperature history by using a new method to capture climate signals from a tree-ring network consisting of 31 sampling sites collected in seven provinces of subtropical China. The objectives of this study were 1) to reconstruct the cold-season temperature history for subtropical China, 2) to reveal frequency changes in the occurrence of exceptional cold-season temperature decreases over the past two centuries, and 3) to identify the main driver for the variation in cold-season temperatures in subtropical China.

2. Materials and methods

a. Tree-ring data

Tree-ring samples were collected from 31 sites of old growth Pinus (hereafter indicated by P.) massoniana and P. taiwanensis forests in seven provinces of southern China (i.e., south of Qinling Mountain and Huaihe River), covering an area of 94.6 × 104 km2 (from approximately 24° to 33°N and 109° to 122°E) in the subtropical region (Fig. 1 and Table 1). For each site, at least 30 old trees were chosen for sampling. Increment core samples (one core per tree) were collected at breast height from each tree and in a direction parallel to the mountain's slope. In the laboratory, the increment core samples were mounted and polished to make the tree rings clearly visible. Tree-ring widths for each sample were measured (to a precision of 0.001 mm) using a tree-ring measurement system (Lintab, Germany). The measured tree-ring sequences were cross dated following standard dendrochronological procedures (Fritts 1976; Cook et al. 1990). Samples that could not be cross dated because of poor quality (such as containing too many rotten or broken pieces) were removed from further analyses. A total of 838 samples were successfully cross dated and used for analysis in this study. The lowest number of samples for a site is 15 trees, and the highest is 44 trees (Table 1). Among the 31 sampling sites, the tree-ring data for 26 sites are first presented in this study and only five sites were used in our previous study (Duan et al. 2012).

Fig. 1.

Map of the spatial distribution of the 31 tree-ring sampling sites and the 73 meteorological stations. Seven different provinces are denoted with different colors.

Fig. 1.

Map of the spatial distribution of the 31 tree-ring sampling sites and the 73 meteorological stations. Seven different provinces are denoted with different colors.

Table 1.

Descriptions of the 31 tree-ring sampling sites in subtropical China. In the Species column, P. refers to Pinus.

Descriptions of the 31 tree-ring sampling sites in subtropical China. In the Species column, P. refers to Pinus.
Descriptions of the 31 tree-ring sampling sites in subtropical China. In the Species column, P. refers to Pinus.

b. Climatic data

Climate data from 73 meteorological stations around the 31 sampling sites were used in this study (Fig. 1). The common period of these climate data is from 1957 to 2008. Monthly mean temperature and monthly total precipitation are of interest in this study. The first principal components (PC1) of monthly mean temperature and monthly total precipitation for each month were extracted from the 73 meteorological stations over the period 1957–2008 to represent the large-scale climate conditions. Variance explained by the PC1 of monthly mean temperature ranges from 54.5% (August) to 89.4% (February) and is greater than 77.5% for each month from January to April. For the mean temperature in January–March, the variance explained by PC1 is 86.2%. For monthly total precipitation, the variance explained by PC1 of each month ranges from 22.5% to 62.4% and is 48.7% for January–March. The correlation coefficients of monthly temperature for each month (especially from January to March) among the 73 meteorological stations are mostly greater than 0.85 over the period 1957–2008. Because the PC1 of January–March mean temperature explained 86.2% of the total variance and was highly correlated (r ≈ 1) with the averaged January–March mean temperature of the 73 stations, the latter was used for establishing a transfer function to reconstruct past climate.

c. Ring-width growth decrease analysis

Ring-width growth changes (RGC) for each growth year relative to the previous year were calculated for each sampled tree using the following equation in the list ring measurement (LRM) program (Grissino et al. 1993):

 
formula

where rt refers to the ring width in year t and rt−1 refers to the ring width in year t − 1. A negative growth change of greater than 20% was considered to be an anomalous ring-width growth decrease (RGD) and a severe deviation from the natural growth trend (Zhang and Alfaro 2002). The percentage of trees having negative anomalous growth change (i.e., PRGD) was calculated for each year from 1772 to 2008 using the data from all 838 tree-ring samples. As the sample size declined back in time, the PRGD series were truncated when the sample size dropped below 45 trees. Relationships between PRGD and monthly climate factors were examined by Pearson's correlation analysis in the period 1958–2008. In addition, the correlation between PRGD and the interannual difference in monthly climate (i.e., the value in the current year minus that of the previous year) was also calculated.

d. Calculation of the SHI

The Siberian high intensity index (SHI) was calculated as a regionally (40°–60°N, 80°–120°E) averaged sea level pressure (SLP), which was defined by Wu and Wang (2002). The SLP data were obtained from the National Oceanic and Atmospheric Administration/Office of Oceanic and Atmospheric Research/Earth System Research Laboratory/Physical Science Division (NOAA/OAR/ESRL/PSD; http://www.esrl.noaa.gov/psd/). The data have a spatial resolution of 2° latitude by 2° longitude and cover the period 1871–2008.

e. Climate reconstruction

The interannual difference in January–March mean temperature (IADT1–3) was reconstructed from PRGD series using the linear regression method. The skill of the regression model was tested using both the split calibration/verification (Meko and Graybill 1995) and the leave-one-out cross-validation methods (Michaelsen 1987). The statistics for evaluation of the regression model include Pearson's correlation coefficient R, reduction of error (RE), sign test (ST), product mean test (PMT), and coefficient of efficiency (CE). Both RE and CE are measures of shared variance between actual and estimated series, with a positive value suggesting that the reconstruction has encouraging performance (Cook et al. 1994).

3. Results

a. The tree-ring PRGD series

Based on the tree-ring width network consisting of samples from 838 trees collected from 31 sites, the percentage of trees in each year having a tree-ring width decrease greater than 20% relative to the previous year was calculated. The resultant series covered the period 1825–2008. The first year (i.e., 1825) corresponds to a minimum sample size of 45 trees from 11 sampling sites (Fig. 2).

Fig. 2.

The PRGD for sample sizes greater than 45 back to year 1825.

Fig. 2.

The PRGD for sample sizes greater than 45 back to year 1825.

b. Growth–climate relationships

Correlation analysis showed that the PRGD series was significantly correlated with regional temperature (i.e., PC1) in January (R = −0.34, n = 51, p < 0.05), February (R = −0.35, n = 51, p < 0.05), and January–March (R = −0.43, n = 51, p < 0.01), but no significant correlation was found between PRGD and precipitation (Fig. 3a). Further examination showed that the PRGD series had a stronger correlation with the interannual difference of January–March mean temperature (IADT1–3; R = −0.70, n = 51, p < 0.0001) (Fig. 3b).

Fig. 3.

Correlation coefficients between PRGD and (a) climate factors (PC1 of the 73 meteorological stations) and (b) the interannual difference of climate factors from the previous October to the current September and January–March means over the period 1958–2008. In the figure, “Mt” and “Pre” represent monthly mean temperature and monthly precipitation, respectively; “py” means previous year; “J-M” means January–March; and horizontal dotted lines indicate the statistical significance level at p = 0.05.

Fig. 3.

Correlation coefficients between PRGD and (a) climate factors (PC1 of the 73 meteorological stations) and (b) the interannual difference of climate factors from the previous October to the current September and January–March means over the period 1958–2008. In the figure, “Mt” and “Pre” represent monthly mean temperature and monthly precipitation, respectively; “py” means previous year; “J-M” means January–March; and horizontal dotted lines indicate the statistical significance level at p = 0.05.

c. Reconstruction of interannual January–March temperature change

According to the growth–climate relationship (Fig. 3), we produced a well-validated and absolutely dated 184-yr (1825–2008) reconstruction of cold-season temperature change (i.e., IADT1–3; Fig. 4a). Calibration and verification tests indicated good performance of the regression model (Table 2). The values of R, R2, and F in the early, late, and full period were all significant at the level of p < 0.01. Durbin–Watson (DW) statistics obtained from the early, late, and full periods (2.52, 2.10, and 2.41) denoted no first-order autocorrelation in the model residuals. Positive RE and CE statistics in each verification period indicated that the regression model exhibits good skill in reconstructing past climate (Fritts 1976; Cook et al. 1994). The PMT values, a measure of the sign and magnitude of departure from the calibration mean, were also significant at the 0.01 level. The sign test results were significant at the level p < 0.01, indicating that the tree rings tracked the direction of change in climate from year to year well (Pederson et al. 2001). The reconstructed temperature series agreed well with the actual temperature and explained 49.5% of the variance in instrumental climate in the period 1958–2008 (Fig. 4a).

Fig. 4.

Characteristics of IADT1–3 over the period 1825–2008. (a) Reconstructed interannual January–March temperature changes (IADT1–3) back to the year 1825 and the comparison with instrumental IADT1–3 during the period 1958–2008. (b) Years of exceptional interannual temperature decrease indicated by PRGD > 40%. (c) Shift in mean for our previous January–April temperature reconstruction (Duan et al. 2012) (probability = 0.05 and cutoff length = 10), where the red line indicates the regime shift in the mean value. (d) Winter half-year temperature anomaly in eastern China (approximately 25°–40°N and 105°–120°E) with 10-yr resolution during 1825–1995 (Ge et al. 2003).

Fig. 4.

Characteristics of IADT1–3 over the period 1825–2008. (a) Reconstructed interannual January–March temperature changes (IADT1–3) back to the year 1825 and the comparison with instrumental IADT1–3 during the period 1958–2008. (b) Years of exceptional interannual temperature decrease indicated by PRGD > 40%. (c) Shift in mean for our previous January–April temperature reconstruction (Duan et al. 2012) (probability = 0.05 and cutoff length = 10), where the red line indicates the regime shift in the mean value. (d) Winter half-year temperature anomaly in eastern China (approximately 25°–40°N and 105°–120°E) with 10-yr resolution during 1825–1995 (Ge et al. 2003).

Table 2.

Statistics of the calibration and verification for the regression model using tree rings as the independent and climate as the dependent variable. Statistics R, R2, R2adj, F, ST, and PMT are all significant at the level of p < 0.01.

Statistics of the calibration and verification for the regression model using tree rings as the independent and climate as the dependent variable. Statistics R, R2, R2adj, F, ST, and PMT are all significant at the level of p < 0.01.
Statistics of the calibration and verification for the regression model using tree rings as the independent and climate as the dependent variable. Statistics R, R2, R2adj, F, ST, and PMT are all significant at the level of p < 0.01.

d. Characteristics of interannual January–March temperature variations

In the PRGD series, there are 23 years that the PRGD values are greater than 40%. We found that 10 out of the 23 anomalous years fall within the instrumental period (1958–2008), and the values of IADT1–3 in these 10 years almost exceed 1.2 standard deviations (SD) of the period 1958–2008 (which is 1.27°C) (Table 3). Therefore, the years with a PRGD greater than 40% can be identified as exceptional years of interannual change in January–March temperature. It is noteworthy that the 23 years with PRGD > 40% are unevenly distributed over the period 1825–2008, occurring at a significant frequency after the year 1935 (χ2 = 6.83, p < 0.01) (Fig. 4b).

Table 3.

Years of PRGD > 40% that are validated by instrumental records of low interannual difference in January–March temperature (IADT1–3) (1.2SD = 1.27°C) or phenological documents of delayed days in first flowering.

Years of PRGD > 40% that are validated by instrumental records of low interannual difference in January–March temperature (IADT1–3) (1.2SD = 1.27°C) or phenological documents of delayed days in first flowering.
Years of PRGD > 40% that are validated by instrumental records of low interannual difference in January–March temperature (IADT1–3) (1.2SD = 1.27°C) or phenological documents of delayed days in first flowering.

e. Relationship between IADT1–3 and SHI

The observed IADT1–3 presented a strong and inverse correlation with SHI (R = −0.78, n = 51) in the instrumental period, 1958–2008 (Fig. 5a). The same relationship existed for our reconstructed IADT1–3 series (R = −0.52, n = 51) (Fig. 5b). Interestingly, such a significant correlation only remained in the period 1935–2008 (R = −0.53, n = 74, p < 0.0001) and declined backwards for the period 1872–1934 (R = −0.21, n = 63, p = 0.103) (Fig. 5c). Correlation coefficients over a moving 50-yr window also showed that a significant correlation did not exist before the mid-1930s (Fig. 5d).

Fig. 5.

Relationships between IADT1–3 and SHI. Comparison of the interannual difference of January–March SHI with the interannual difference of (a) instrumental January–March temperature (R = −0.78, 1958–2008), (b) instrumental-period reconstruction of January–March temperature (R = −0.52, 1958–2008), and (c) whole-period (covered by SHI data) reconstruction of January–March temperature (R = −0.21, 1872–1934; R = −0.53, 1935–2008). (d) Correlation between SHI and IADT1–3 in a moving 50-yr window over the period 1872–2008. Note that the SHI series in (a)–(c) were all multiplied by −1. The Z scores represent the standardized series.

Fig. 5.

Relationships between IADT1–3 and SHI. Comparison of the interannual difference of January–March SHI with the interannual difference of (a) instrumental January–March temperature (R = −0.78, 1958–2008), (b) instrumental-period reconstruction of January–March temperature (R = −0.52, 1958–2008), and (c) whole-period (covered by SHI data) reconstruction of January–March temperature (R = −0.21, 1872–1934; R = −0.53, 1935–2008). (d) Correlation between SHI and IADT1–3 in a moving 50-yr window over the period 1872–2008. Note that the SHI series in (a)–(c) were all multiplied by −1. The Z scores represent the standardized series.

4. Discussion

a. Climate signal from tree rings in subtropical China

In this study, January–March temperature was found to be the climate factor affecting tree-ring growth in large-scale subtropical China. Low January–March temperature might cause a delay to the beginning of tree radial growth if the coldness extends further into the spring (Zhu et al. 2009; Duan et al. 2012). Severe cold conditions might also cause bud damage, frost desiccation, and fine root mortality (Körner 1998; Pederson et al. 2004; Fan et al. 2008), all resulting in a relatively narrow tree ring in the growing season. Conversely, warm winters could allow trees to synthesize nonstructural carbohydrates and other organic substances, enhancing early wood growth in the growing season (Chabot and Hicks 1982). Such a cold-season temperature constraint on tree-ring growth was also reported in other studies in temperate and subtropical regions (Pederson et al. 2004; Yonenobu and Eckstein 2006; Zhu et al. 2009; Shi et al. 2010; Duan et al. 2012).

The extraction of climate signals from tree rings in subtropical China has been a challenge because of the weak growth–climate relationship in single-site chronology (Duan et al. 2012; Shi et al. 2010). In this study, a strong climate signal was obtained from our tree-ring network using a new method examining the percentage of trees in each year that show abnormal growth reduction. The resultant series of tree percentage was a good indicator of large-scale climatic change. The effect of decreasing sample size on climate reconstruction was tested and the results showed no significant influence on the pattern of climate reconstruction using the lowest number of sample size in this study. The strongest climate signal is IADT1–3 rather than the temperature in the growth year, indicating that the anomalous ring-width growth change is more sensitive to change in climate condition (i.e., interannual difference of climate) than state of climate in the growth year (Figs. 3a,b).

The 23 exceptional IADT1–3 events can be validated by observed climate records or phenological documents. All of the 10 years with PRGD > 40% in the instrumental period were demonstrated to be years of temperature decrease and 7 of the 10 years showed temperature decreases exceeding 1.2 SD (Table 3). In the preinstrumental period, the phenological record of plants flowering in the Hunan province in the period 1888–1916 helps validation of the tree-ring reconstructed cold-season temperature decrease (Fang et al. 2005). Three years with PRGD > 40% (i.e., 1916, 1905, and 1891) in the period 1888–1916 corresponded with the years of delayed flowering (all delayed more than 12 days) (Table 3). An anomalously cold January–March could induce intense and extended soil freezing and thus delay the effective beginning of spring and plant flowering (Dunne et al. 2003).

b. Potential driver for the increased variability of cold-season temperature after the 1930s

The more frequent exceptional values of PRGD (>40%) since the mid-1930s suggests an increased variability in cold-season temperature in subtropical China (Fig. 4b). Our previous reconstruction of cold-season temperatures for two provinces of this region also showed an increased number of regime shifts since the 1930s (Fig. 4c). What drove such increased variability of cold-season temperature after the 1930s?

The variability of East Asia's winter climate is complex, involving multiple influencing factors such as the East Asian winter monsoon (Zhu 2008; Wang et al. 2010), Arctic Oscillation (Chen and Zhou 2012; Gong et al. 2004; Huang et al. 2007), and Siberian high (Wen and Chen 2006; Li et al. 2007; Zhang et al. 2009). Among these forcings, the Siberian high seems to play a dominant role in affecting anomalous cold-season temperature change in subtropical China (Wen and Chen 2006; Li et al. 2007; Ding et al. 2008; Zhang et al. 2009). For the cold extreme in January 2008, a persistent anomaly of atmospheric circulation in Eurasia led a branch of the strong high-level westerly to stretch northward and then turn southward and was considered to be the direct cause for the succession of cold air incursion (Ding et al. 2008). In our study, both the strong correlation between instrumental and reconstructed IADT1–3 and the Siberian high intensity index (SHI) in the instrumental period indicated the dominant driving effect of Siberian high on the interannual cold-season temperature change in subtropical China (Figs. 5a,b). However, our 184-yr tree-ring record showed that the tight relationship remained only back to the mid-1930s, suggesting that the dominant driving effect was unstable on a long time scale (Figs. 5c,d). This unstable relationship might be related to the climate's warming.

We found that the increased variability in cold-season temperature since the mid-1930s in subtropical China corresponds to a warm interval in winter half-year temperatures in eastern China (which includes our study area) (Ge et al. 2003) (Fig. 4d). In fact, such a temperature increase was also found in Europe and the Northern Hemisphere. An increase in temperature since the 1930s was reported in west-central Scandinavia (Gunnarson et al. 2011). Instrumental records of Northern Hemispheric temperatures back to the past two to three centuries also presented a positive temperature anomaly since about the 1930s (Jansen et al. 2007). Studies of the recent cold and snowy winters in Europe and northeastern and midwestern United States showed that the decrease in autumn Arctic sea ice area from the climate's warming resulted in changes in the Northern Hemisphere's winter atmospheric circulation and caused much broader meridional meanders at midlatitudes and different interannual variability (Liu et al. 2012). Another study of instrumental records about cold wave frequency (CWF) indicated that the primary response of CWF to a warming climate may be the southward shift of the cold center (Ma et al. 2012). The model results showed that southern China tends to experience more cold waves than northern China in the twenty-first century (2045–64 and 2080–99) under global warming (Ma et al. 2012). These phenomena suggest that the increased variability in interannual change of cold-season temperature since the mid-1930s might be related to the climate's warming.

The findings of this study indicate that, despite global warming, subtropical China could experience abnormal decreases in winter temperatures and such temperature decreases have occurred frequently since the 1930s rather than only in recent years. Our results suggest that global warming strengthened the influence of the Siberian high on cold-season temperature change in subtropical China and such an influence may continue in the course of the climate's warming. This strengthened association provides insights into the linkage between high-latitude climate systems and low-latitude climate.

Acknowledgments

This study was supported by the National Natural Science Foundation of China (Grants 40631002, 31170419, and 41101043) and China Postdoctoral Science Foundation (Grant 20100480522). Jianping Duan gratefully acknowledges the support of the K. C. Wong Education Foundation. Climate data from the meteorological stations were obtained from the National Meteorological Information Center of the China Meteorological Administration. We thank Dr. Chao Zhang, Dr. Lushuang Gao, Ms. Gaiai Guo, and Ms. Hongyan Qiu for assisting in field work and Caiyun Liu and Hongyan Qiu for assisting in tree-ring cross dating.

REFERENCES

REFERENCES
Cameron
,
I.
,
2012
:
Snow patches in England and Wales during spring 2012
.
Weather
,
67
,
247
247
.
Chabot
,
B. F.
, and
D. J.
Hicks
,
1982
:
The ecology of leaf life spans
.
Annu. Rev. Ecol. Syst.
,
13
,
229
259
.
Chen
,
W.
, and
Q.
Zhou
,
2012
:
Modulation of the Arctic oscillation and the East Asian winter climate relationships by the 11-year solar cycle
.
Adv. Atmos. Sci.
,
29
,
217
226
.
Cook
,
E. R.
,
S.
Shiyatov
, and
V.
Mazepa
,
1990
: Estimation of the mean chronology. Methods of Dendrochronology: Applications in the Environmental Sciences, E. R. Cook and L. A. Kairiukstis, Eds., Kluwer Academic, 123–132.
Cook
,
E. R.
,
K. R.
Briffa
, and
P. D.
Jones
,
1994
:
Spatial regression methods in dendroclimatology—A review and comparison of 2 techniques
.
Int. J. Climatol.
,
14
,
379
402
.
Ding
,
Y. H.
,
Z. Y.
Wang
,
Y. F.
Song
, and
J.
Zhang
,
2008
:
The unprecedented freezing disaster in January 2008 in southern China and its possible association with the global warming
.
Acta. Meteor. Sin.
,
22
,
538
558
.
Duan
,
J. P.
,
Q. B.
Zhang
,
L. X.
Lv
, and
C.
Zhang
,
2012
:
Regional-scale winter-spring temperature variability and chilling damage dynamics over the past two centuries in southeastern China
.
Climate Dyn.
,
39
,
919
928
.
Dunne
,
J. A.
,
J.
Harte
, and
K. J.
Taylor
,
2003
:
Subalpine meadow flowering phenology responses to climate change: Integrating experimental and gradient methods
.
Ecol. Monogr.
,
73
,
69
86
.
Fan
,
Z. X.
,
A.
Brauning
, and
K. F.
Cao
,
2008
:
Annual temperature reconstruction in the central Hengduan Mountains, China, as deduced from tree rings
.
Dendrochronologia
,
26
,
97
107
.
Fang
,
X. Q.
,
L. B.
Xiao
,
Q. S.
Ge
, and
J. Y.
Zheng
,
2005
:
Change of plants phenophases and temperature in spring during 1888–1916 around Changsha and Hengyang in Hunan province (in Chinese)
.
Quat. Sci.
,
25
,
74
79
.
Fritts
,
H. C.
,
1976
: Tree Rings and Climate. Academic Press, 567 pp.
Ge
,
Q. S.
,
J. Y.
Zheng
,
X. Q.
Fang
,
Z. M.
Man
,
X. Q.
Zhang
,
P. Y.
Zhang
, and
W. C.
Wang
,
2003
:
Winter half-year temperature reconstruction for the middle and lower reaches of the Yellow River and Yangtze River, China, during the past 2000 years
.
Holocene
,
13
,
933
940
.
Gong
,
D. Y.
,
S. W.
Wang
, and
J. H.
Zhu
,
2004
:
Arctic oscillation influence on daily temperature variance in winter over China
.
Chin. Sci. Bull.
,
49
,
637
642
.
Grissino
,
M. H.
,
R.
Holmes
, and
H. C.
Fritts
,
1993
: International tree-ring data bank program library user's manual. University of Arizona Laboratory of Tree-Ring Research Rep., 76 pp.
Gunnarson
,
B. E.
,
H. W.
Linderholm
, and
A.
Moberg
,
2011
:
Improving a tree-ring reconstruction from west-central Scandinavia: 900 years of warm-season temperatures
.
Climate Dyn.
,
36
,
97
108
.
Huang
,
J. Y.
,
B. K.
Tan
,
L. L.
Suo
, and
Y. Y.
Hu
,
2007
:
Monthly changes in the influence of the Arctic oscillation on surface air temperature over China
.
Adv. Atmos. Sci.
,
24
,
799
807
.
Jansen
,
E. J.
, and
Coauthors
,
2007
: Palaeoclimate. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 433–497.
Körner
,
C.
,
1998
:
A re-assessment of high elevation treeline positions and their explanation
.
Oecologia
,
115
,
445
459
.
Li
,
Y.
,
2012
:
Analysis of the January 2012 atmospheric circulation and weather (in Chinese)
.
Meteor. Monogr.
,
38
,
495
500
.
Li
,
Y.
,
R. Y.
Lu
, and
J. H.
He
,
2007
:
Several climate factors influencing the winter temperature over China (in Chinese)
.
Chin. J. Atmos. Sci.
,
31
,
506
514
.
Liu
,
C. F.
, and
B.
Gao
,
2001
:
Climate of chilling damage in spring in southern China and its atmospheric circulation features (in Chinese)
.
J. Trop. Meteor.
,
17
,
179
187
.
Liu
,
J. P.
,
J. A.
Curry
,
H. J.
Wang
,
M. R.
Song
, and
R. M.
Horton
,
2012
:
Impact of declining Arctic sea ice on winter snowfall
.
Proc. Nat. Acad. Sci. USA
,
109
,
6781
6783
.
Ma
,
T. T.
,
Z. W.
Wu
, and
Z. H.
Jiang
,
2012
:
How does coldwave frequency in China respond to a warming climate?
Climate Dyn.
,
39
,
2487
2496
.
Meko
,
D.
, and
D. A.
Graybill
,
1995
:
Tree-ring reconstruction of upper Gila River discharge
.
Water Resour. Bull.
,
31
,
605
616
.
Michaelsen
,
J.
,
1987
:
Cross-validation in statistical climate forecast models
.
J. Climate Appl. Meteor.
,
26
,
1589
1600
.
Moore
,
G. W. K.
, and
I. A.
Renfrew
,
2012
:
Cold European winters: Interplay between the NAO and the east Atlantic mode
.
Atmos. Sci. Lett.
,
13
,
1
8
.
Pederson
,
N.
, and
Coauthors
,
2001
:
Hydrometeorological reconstructions for northeastern Mongolia derived from tree rings: 1651–1995
.
J. Climate
,
14
,
872
881
.
Pederson
,
N.
,
E. R.
Cook
,
G. C.
Jacoby
,
D. M.
Peteet
, and
K. L.
Griffin
,
2004
:
The influence of winter temperatures on the annual radial growth of six northern range margin tree species
.
Dendrochronologia
,
22
,
7
29
.
Ratnam
,
J. V.
,
S. K.
Behera
,
Y.
Masumoto
,
K.
Takahashi
, and
T.
Yamagata
,
2012
:
Anomalous climatic conditions associated with the El Niño Modoki during boreal winter of 2009
.
Climate Dyn.
,
39
,
227
238
.
Seager
,
R.
,
Y.
Kushnir
,
J.
Nakamura
,
M.
Ting
, and
N.
Naik
,
2010
:
Northern Hemisphere winter snow anomalies: ENSO, NAO and the winter of 2009/10
.
Geophys. Res. Lett.
,
37
,
doi:10.1029/2010GL043830
.
Shi
,
J. F.
,
E. R.
Cook
,
H. Y.
Lu
,
J. B.
Li
,
W. E.
Wright
, and
S. F.
Li
,
2010
:
Tree-ring based winter temperature reconstruction for the lower reaches of the Yangtze River in southeast China
.
Climate Res.
,
41
,
169
175
.
Wang
,
B.
,
Z. W.
Wu
,
C. P.
Chang
,
J.
Liu
,
J. P.
Li
, and
T. J.
Zhou
,
2010
:
Another look at interannual-to-interdecadal variations of the East Asian winter monsoon: The northern and southern temperature modes
.
J. Climate
,
23
,
1495
1512
.
Wen
,
K. G.
, and
S. X.
Chen
,
2006
: China Meteorological Calamity Grand Ceremony: Jiangxi Volume (in Chinese). Meteorological Press, 470 pp.
Wu
,
B.
, and
J.
Wang
,
2002
:
Winter Arctic Oscillation, Siberian high and East Asian winter monsoon
.
Geophys. Res. Lett.
,
29
,
2002
,
doi:10.1029/2002GL015373
.
Yonenobu
,
H.
, and
D.
Eckstein
,
2006
:
Reconstruction of early spring temperature for central Japan from the tree-ring widths of Hinoki cypress and its verification by other proxy records
.
Geophys. Res. Lett.
,
33
,
L10701
,
doi:10.1029/2006GL026170
.
Zhang
,
Q. B.
, and
R. I.
Alfaro
,
2002
:
Periodicity of two-year cycle spruce budworm outbreaks in central British Columbia: A dendro-ecological analysis
.
For. Sci.
,
48
,
722
731
.
Zhang
,
Z. Y.
,
D. Y.
Gong
,
M.
Hu
,
D.
Guo
,
X. Z.
He
, and
Y. N.
Lei
,
2009
:
Anomalous winter temperature and precipitation events in southern China
.
J. Geogr. Sci.
,
19
,
471
488
.
Zhao
,
L.
, and
Coauthors
,
2008
:
Disasters and its impact of a severe snow storm and freezing rain over southern China in January 2008 (in Chinese)
.
Climate Environ. Res.
,
13
,
556
566
.
Zhu
,
H. F.
,
X. Q.
Fang
,
X. M.
Shao
, and
Z. Y.
Yin
,
2009
:
Tree ring-based February-April temperature reconstruction for Changbai Mountain in northeast China and its implication for East Asian winter monsoon
.
Climate Past
,
5
,
661
666
.
Zhu
,
Y. F.
,
2008
:
An index of East Asian winter monsoon applied to the description of China's mainland winter temperature changes
.
Acta. Meteor. Sin.
,
22
,
522
529
.