Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Month-to-Month Variability of Autumn Sea Ice in the Barents and Kara Seas and Its Relationship to Winter Air Temperature in China

Month-to-Month Variability of Autumn Sea Ice in the Barents and Kara Seas and Its Relationship to... Hindawi Advances in Meteorology Volume 2019, Article ID 4381438, 13 pages https://doi.org/10.1155/2019/4381438 Research Article Month-to-Month Variability of Autumn Sea Ice in the Barents and Kara Seas and Its Relationship to Winter Air Temperature in China 1 1 2 3 1 Chuhan Lu , Kaili Li, Shaoqing Xie, Zhaomin Wang, and Yujing Qin Key Laboratory of Meteorological Disaster, Ministry of Education, Joint International Research Laboratory of Climate and Environment Change, Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing 210044, China Jiashan Meteorological Bureau, Jiaxing 314100, China College of Oceanography, Hohai University, Nanjing 210000, China Correspondence should be addressed to Yujing Qin; qinyujing@nuist.edu.cn Received 27 May 2019; Revised 30 August 2019; Accepted 29 October 2019; Published 12 December 2019 Academic Editor: Alastair Williams Copyright © 2019 Chuhan Lu et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +e variation of autumn Arctic sea ice is a critical indicator of temperature anomalies over the Eurasian continent during winter. +e retreat of autumn Arctic sea ice is typically accompanied by negative anomalous winter temperatures over the Eurasian and North American continents. However, such sea ice temperature linkages notably change from month to month. +e variation of the autumn Arctic sea ice area and the relationship between the month-to-month sea ice and winter temperature anomalies in China are investigated using the Hadley Centre’s sea ice dataset (HadiSST) and the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis dataset (ERA-Interim) during 1979–2018. We present the following results: +e sea ice in the Barents and Kara seas (BK) during the autumn and winter seasons shows notable low-frequency variability. +e retreat of sea ice in the BK from September to November is significantly associated with negative temperature anomalies in the following winter in China. However, the linkage between the sea ice in the BK in September and the winter temperatures is stronger than that in both October and November. An anomalous positive surface pressure is exhibited over the northwestern part of Eurasia in the winter that is linked to decreasing sea ice in the BK in the preceding September. +is surface pressure favors the persistence and intensification of synoptic perturbations, such as blocking highs and surface cold highs, as well as the intensification of the Siberian High and the East Asian winter monsoon. +ese favorable conditions ultimately contribute to the formation of large-scale winter cold anomalies in China. Compared to low sea ice cover in October and November, a more oceanic heat storage in the upper BK induced by low sea ice cover in the BK leads to a larger heat release to tropospheric atmosphere in winter by surface heat flux and upward longwave radiation in the BK. +is regional tropospheric warming results in a higher barotropic positive height anomaly over the Ural Mountains, and then more active cold advection from the high latitude affects East Asia. [2, 3]. Along with this unforeseen cooling trend in the 1. Introduction midlatitude regions of the NH, both the Cold Siberian High Arctic sea ice is a prominent indicator of global climate and the East Asian winter monsoon exhibit a substantial change. A substantial loss of ice and a higher warming rate decadal intensification, leading to more active cold air have been observed in the Arctic during recent decades, a outbreaks in China [4, 5]. Recent studies documented that phenomenon known as Arctic amplification, with a con- there may be some physical linkages between the recent tinuous increase in the global air temperature [1]. Severe reduction of ice in the Arctic Ocean and cooling in the cold events and anomalously heavy snowfalls have fre- midlatitude continents of the NH through the anomalous quently affected most midlatitude regions of the Northern atmospheric circulation response of sea ice loss [6]. For Hemisphere (NH), including major industrialized centers example, a reduction in the Arctic sea ice is typically 2 Advances in Meteorology accompanied by substantial warming that weakens the reduction that occurs east of the Arctic Ocean and the BK, is meridional temperature gradient. +is will decelerate the associated with substantial surface warming. A weakening westerlies in the midlatitudes and result in a decrease in polar/midlatitude temperature gradient results in a Eurasian atmospheric baroclinicity and a more wavy circulation [7, 8]. (EU) pattern response that facilitates the strengthening of Compared with the direct response within the troposphere, the Siberian High and the East Asian winter monsoon, stratospheric response to sea ice forcing also plays a robust which brings substantial cooling and more extreme low role in the development of cold conditions over midlatitude events in East Asia [22–24]. Li et al. [25] documented the continents of the NH [9]. +ere is still a debate regarding the local responses of the temperature in East Asia to changes in physical connection between the changes in Arctic sea ice sea ice in different regions. +ey reported that the retreat of and the midlatitude activity of planetary waves [10]. A the sea ice in the BK induces an intensification of the change in the dominant mode of extratropical circulation Siberian High and lowers the temperature in the area north may also have a close association with the recent cooling of East Asia, while an increase in sea ice in the Okhotsk Sea trend [11–13]. induces a northerly shift of the jet stream in the upper Previous studies have shown that extratropical impacts troposphere over East Asia with substantial warming in the depend highly on the regional structure of the anomalous area south of East Asia. +e variations of the sea ice in the BK Arctic climate state [14]. For example, warming over the BK are associated with the Arctic Oscillation (Arctic Dipole region can lead to East Asian cooling, whereas northern pattern) by inducing cross-Arctic-like (Eurasian) telecon- North America cooling is closely related to warming over the nection wave-train patterns that lead to an intensification of East Siberian Sea–Chukchi Sea region [14]. +e reduction of cold surge activity in China [26]. +e SST anomalies in the sea ice over the BK also has a cooling effect on the European NH and the changes in Arctic sea ice were involved in the region induced by more blockings associated with the loss of interdecadal changes in the winter surface air temperature sea ice in the BK [15]. However, atmospheric responses to (SAT) over East Asia that occurred approximately during the sea ice variations of the BK vary with background at- the mid-1990s [4, 27]. mospheric states and seasons. A recent study had pointed It is worthy to note that the variation of the sea ice cover shows clear seasonal and regional dependencies [28], in out that the loss of sea ice in the BK in November links with a weaker SH in November, while a stronger SH in December spite of a general decrease trend of the Arctic sea ice. [16]. Furthermore, the relationships between wintertime SAT Many previous studies have focused on the relationship anomalies in China and sea ice variation in different time between the variations in Arctic sea ice and atmospheric scales or regions also display notable diversity [23, 29]. circulation patterns and the associated anomalous climate in +erefore, we will study the association of month-to- China. Xie and Huang [17] reported that changes in Arctic month variability of the regional Arctic sea ice cover in the sea ice and sea surface temperature (SST) anomalies in the preceding autumn with the SAT anomalies and its possible central and eastern equatorial Pacific play important roles in linking pathway. the general circulation and suggested that there may be physical connections between them. Wu et al. [18] con- 2. Data and Methods firmed that the variations in the areas of sea ice in the Greenland Sea and BK are associated with the occurrence of 2.1. Data. +e data used in this study include the following: El Niño-Southern Oscillation (ENSO) events and that a 3- (1) the European Centre for Medium-Range Weather year leading signal of anomalous sea ice shows a close Forecasts (ECMWF) reanalysis dataset (ERA-Interim) from linkage with the general circulation in the NH. A numerical 1979 to 2018 is used for deriving data on the daily and simulation also indicated that the variation in Arctic sea ice monthly air temperatures, sea-level pressure, geopotential is one of the most important indicators of an East Asian ° ° height, and horizontal wind with a 2.5 × 2.5 horizontal summer monsoon climate anomaly [19]. Shi et al. [20] resolution [30]; (2) monthly mean sea ice concentrations are documented that there is a positive correlation between the derived from the Met Office Hadley Centre’s sea ice and sea zonal temperature gradient of the Arctic Ocean and the surface temperature (HadiSST 1.1) dataset [31]; (3) the geopotential height gradient between the Barents Sea and the monthly mean heat content is derived from the vertical East Siberian Sea at 500 hPa and that the associated average of potential temperature (109.8 m above; 11 levels) anomalous circulation links with the summer precipitation from the ECMWF ocean reanalysis system ORAS4 [32]; and anomalies over Northeast China. Wang et al. [21] also re- (4) observed monthly mean surface air temperatures (SATs) ported that the preceding autumn loss of Arctic sea ice favors are from the National Climate Center of China, including a northward shift of the cyclone track in China along with 160 stations. the weakening of Rossby waves over the area south of 40 N in eastern China, resulting in an intensification of haze weather. 2.2. Circulation Indices. +e definitions of circulation in- +e influences of changes in Arctic sea ice on the dices are as follows: +e westerly index is defined as the wintertime air temperature in China under the background differences in the geopotential heights between 40 N and ° ° ° of Arctic amplification have drawn extensive attention from 65 N and between 0 and 122 E [33]. +e blocking high scholars in recent years. A series of studies showed that the frequency is derived from the anomalous 500 hPa mid- preceding reduction of Arctic sea ice in autumn, such as the latitude geopotential height [34]. +e anticyclone intensity Advances in Meteorology 3 over Eurasia is the seasonal mean of the center pressure for simultaneous correlation coefficients of the OBKI with the each surface high in 6 hourly SLP field during the winter [2]. Arctic sea ice area index (ASI (downloaded from NSIDC ° ° ° ° +e regional mean of the SLP (40 N–60 N, 70 E–120 E) (http://nsidc.org/data/g02135.html))) in September, Oc- represents the intensity of the Cold Siberian High [33], and tober, and November are 0.43, 0.55, and 0.47, respectively, the East Asian winter monsoon index is defined by the which are statistically significant with significance ex- midlatitude land-sea pressure differences between 110 E and ceeding 0.05. +ese significant correlations are suggestive 160 E [35]. In addition, the monthly Arctic Oscillation (AO) of the consistency of the low-frequency sea ice variations index (http://www.cpc.ncep.noaa.gov/products/precip/ between our defined BK region and the Arctic. +e DBKI CWlink/daily_ao_index/ao.shtml) can be downloaded time series shows distinct differences from the OBKI time from the NOAA Climate Prediction Center (CPC). series during heavy and light ice cover years (defined as ± 0.75 σ), as indicated in Figures 2(d)–2(f ). Additional heavy (light) ice cover years occur before (after) the mid- 3. Variability and Linear Trend of the Arctic 1990s in the OBKI time series that bear an important Sea Ice decreasing feature. A further discussion of different as- To show the horizontal distribution of the Arctic sea ice, sociations between the OBKI and the DBKI is presented in the following section. the standard deviation (σ) of the sea ice concentration data from September to December from 1979 to 2018 is shown in Figure 1. Large value areas generally extend from 4. Relationship between the Sea Ice and the the interior Arctic to the high-latitude oceans in the NH Air Temperature and suggest large interannual variability of the sea ice cover in the marginal areas of the Arctic sea ice. However, Both the OBKI and the DBKI exhibit positive correlations clear local characteristics are shown in the above large with the wintertime SAT anomalies in most parts of China, values of σ areas. In particular, large variability of sea ice as shown in Figure 3, despite their inconsistent time evo- ° ° occurs in the 70 N–80 N region of the Barents, Kara, and lutions. As documented in Figures 3(a)–3(c), there are Beaufort seas in September and October (Figures 1 and significant positive correlations between the OBKI and the 1(b)). +e northwest coast of Canada and the east coast of SAT in North China. +is is indicated by a cold anomaly Greenland also display high σ values. As a continuous likely followed by a preceding autumn sea ice loss in the BK, southward expansion of the ice cover, the areas of high σ which is in agreement with the previous studies by Wu et al. shift from the preceding Laptev and Beaufort seas to the [22, 24]. However, the extent of the association decreases Bering Sea and the northeast coast of Canada after No- from September to November. In particular, significant vember (Figures 1(c) and 1(d)). It should be noted that correlations can be observed over Xinjiang and Northeast there is a continuous large variability in the BK from China; the largest area is observed in September, after which September to December. Since the change in sea ice results it shrinks gradually in October and November. +e com- from strong sea ice-atmosphere coupling, high σ values in posite differences of the SAT between the low and high the BK indicate that the interannual/interdecadal varia- OBKI years also show a correlated consistency. In Sep- tion of sea ice in this region may be related to notable tember, low sea ice likely induces a decrease in the SAT in changes in atmospheric and oceanic interactions under North China, with maximum decreases occurring especially the background of Arctic amplification [36]. In addition, below − 1.6 C. In addition, slight cooling is also evident in the variation of the sea ice in the BK can be a good in- Southwest China. +e correlation patterns in October and dicator of the East Asian winter monsoon [24]. To un- November are similar to the September one with a slight derstand the effect of sea ice variability on the winter air reduction of the number of stations with statistical temperature variability in China, we used the time series significance. of the sea ice cover area over the BK (box area in Figure 1) Stronger associations of the DBKI with the SAT in China as the sea ice index (BKI) in this key region. in the autumn are observed in Figures 3(d)–3(f), and the +e original and detrended time series of the BKI, month-to-month correlation variations also decrease similar hereafter referred to as the OBKI and DBKI, respectively, to the association of the OBKI with the SAT. In September, during September-October-November (SON) from 1979 to pronounced areas of larger significance are observed over 2018 are presented in Figure 2. +e pronounced retreat of most of China compared with the OBKI (73.8% stations with autumn sea ice in the BK during the past 40 yrs is indicated significance exceeding 0.05), except for the Tibetan Plateau by the OBKI in Figures 1(a)–1(c) with values of and its vicinity (Figure 3(d)). +e composite differences in 5 2 − 1 5 2 − 1 − 0.81 × 10 the SAT between low and high DBKI years are generally km ·10 yr (September), − 1.36 ×10 km ·10 yr 5 2 − 1 (October), and − 1.19 ×10 km ·10 yr (November), which − 0.8 C in the significant regions. +e significant results are all statistically significant exceeding 0.01. +e different described above represent a good indication of the variation decreasing trends of the BKI in the autumn may be because of the sea ice in the BK in September when compared with the Arctic sea ice areas in different months correspond to the SAT in China in the following winter. In October, dissimilar general circulations [34]. +e remarkable retreat significant SAT anomaly regions are observed in Xinjiang, of autumn sea ice in the BK is evident after 2005, which is Northeast China, North China, and the Changjiang-Huaihe consistent with the accelerated decrease in the area of Arctic River Basin, which decreases in size in November similar to sea ice over the last decade [35]. In addition, the Northeast China. 4 Advances in Meteorology 90°W 90°W 60°W 120°W 60°W 120°W 150°W 30°W 150°W 30°W 0° 180° 0° 180° 150°E 150°E 30°E 30°E 60°E 120°E 60°E 120°E 90°E 90°E 0.1 0.2 0.3 0.4 0.1 0.2 0.3 0.4 (a) (b) 90°W 90°W 60°W 120°W 60°W 120°W 30°W 30°W 150°W 150°W 0° 180° 0° 180° 0.1 150°E 150°E 30°E 30°E 120°E 60°E 120°E 60°E 90°E 90°E 0.1 0.2 0.3 0.4 0.1 0.2 0.3 0.4 (c) (d) Figure 1: Standard deviations of the Arctic sea ice concentration data for September (a), October (b), November (c), and December (d) from 1979 to 2018. +e black dashed box denotes the Barents Sea and the Kara Sea region. To further study the relationship between the sea ice and time series (PC1) are shown in Figures 4(a) and 4(c). +e the SAT, we conducted an empirical orthogonal function EOF1 mode shows a pattern of consistent positive SAT (EOF) analysis for the station SAT anomalies in the winter. departures in most of China (except the Tibetan Plateau) +e variance contribution rates for the first two modes are with a general north-to-south decrease in SAT values. +is 50.7% and 19.3%, which can be well defined according to the indicates either consistent cooling or consistent warming of criterion proposed by North et al. [37]. +e spatial distri- China in the winter. A comprehensive analysis using EOF1 bution of the leading mode (EOF1) and its corresponding and PC1 indicates that a dominant cooling trend emerged in Advances in Meteorology 5 0.8 1.1 0.7 0.6 0.9 0.8 0.5 0.7 0.4 0.6 0.5 0.3 0.4 0.2 0.3 0.2 0.1 0.1 0 0 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 (a) (b) 2.5 1.3 1.2 1.5 1.1 0.5 0.9 0.8 0 0.7 –0.5 0.6 –1 0.5 –1.5 0.4 –2 0.3 0.2 –2.5 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 (c) (d) 2.5 2 1.5 1.5 0.5 –0.5 0.5 –1 –1.5 –0.5 –2 –2.5 –1 –3 –1.5 –3.5 –2 –4 –2.5 –4.5 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 (e) (f) Figure 2: BKI time series for September (a, d), October (b, e), and November (c, f) showing the time series of both the OBKI in the left 5 2 panels (unit: 10 km ) and the DBKI in the right panels. +e straight lines in the left panels denote the linear trends in the OBKI. +ose years with high BKI values (>0.75 SD) and low BKI values (<− 0.75 SD) are marked by circles and black dots, respectively. the mid-1980s and after 2010, while a prominent warming positive correlations between the OBKI, DBKI, and PC1 are trend was dominant in the late 1980s and 2000s with a larger evident, and the correlation between the DBKI and PC1 is SAT departure amplitude in the northern part of China. In clearly higher than that between the OBKI and PC1. A addition, a notable interannual fluctuation can be observed notable correlation decrease is evident from September to since the late 1990s which is confirmed by a wavelet analysis November. Particularly, the correlation coefficient between the DBKI in September and PC1 is 0.48 with significance that shows a substantial 2- to 4-year fluctuation since the early 21st century (figure not shown). exceeding 0.01, while the correlation between the DBKI and To further assess the correlation between the SAT PC1 in November decreases to 0.19. +e significant corre- principal component and the sea ice variation, the corre- lations between the sea ice in the BK and PC1 further suggest lation coefficients between the month-to-month OBKI, that the change in the sea ice in the BK in the autumn can be DBKI, ASI, and PC1 and PC2 were calculated, the results of utilized as a skillful predictor of the winter SAT anomalies in which are documented in Table 1. +e difference between the China, especially for the DBKI in September. However, month-to-month OBKI and DBKI is statistically significant similar decreases in the correlations between the ASI and with significance exceeding 0.01. Moreover, significant PC1 are evident, and the original ASI also shows a lower 0.31 0.26 0.26 0.26 0.31 6 Advances in Meteorology 50°N 50°N 0.05 0.05 0.1 0.26 0.1 0.26 40°N 40°N 0.31 0.31 0 0 0.26 0 0.26 0.31 –0.26 –0.1 –0.1 0.31 –0.4 30°N 30°N –0.05 –0.05 0.26 –0.01 –0.01 20°N 20°N 80°E 100°E 120°E 140°E 80°E 100°E 120°E 140°E –0.3~–0.8 –1.6~–2.4 –0.3~–0.8 –1.6~–2.4 0.3~0.8 1.6~2.4 0.3~0.8 1.6~2.4 –0.8~–1.6 <–2.4 –0.8~–1.6 <–2.4 0.8~1.6 >2.4 0.8~1.6 >2.4 (a) (b) 50°N 50°N 0.05 0.05 0.26 0.1 0.1 0.31 40°N 40°N 0.4 0.4 0 0.31 0 0.26 0.31 0.31 0.26 –0.1 –0.1 0.4 30°N 30°N –0.05 –0.05 –0.01 –0.01 20°N 20°N 80°E 100°E 120°E 140°E 80°E 100°E 120°E 140°E –0.3~–0.8 –1.6~–2.4 –0.3~–0.8 –1.6~–2.4 0.3~0.8 1.6~2.4 0.3~0.8 1.6~2.4 –0.8~–1.6 <–2.4 –0.8~–1.6 <–2.4 0.8~1.6 >2.4 0.8~1.6 >2.4 (c) (d) 50°N 0.05 50°N 0.05 0.26 0.31 0.31 0.1 0.1 40°N 0.31 40°N 0.31 0.26 0 0 0.26 0.31 –0.1 –0.1 30°N 30°N –0.05 –0.05 0.26 –0.01 –0.01 20°N 20°N 80°E 100°E 120°E 140°E 80°E 100°E 120°E 140°E –0.3~–0.8 –1.6~–2.4 –0.3~–0.8 –1.6~–2.4 0.3~0.8 1.6~2.4 0.3~0.8 1.6~2.4 –0.8~–1.6 <–2.4 –0.8~–1.6 <–2.4 0.8~1.6 >2.4 0.8~1.6 >2.4 (e) (f) Figure 3: Correlation coefficients between the BKI for September (a, d), October (b, e), and November (c, f) and the winter (DJF) temperature time series for 160 stations in China (shaded: significance and contour: correlation coefficient values) and the composite differences in the temperatures (dots; unit: C) between selected low BKI years and high BKI years. +e upper panels are for the OBKI, and the bottom panels are for the DBKI. Both the shaded areas and the dots indicate values that are statistically significant with significance exceeding 0.1 based on the results of t-tests. +e values in the color bar represent the significance levels of the t-test. positive correlation with the SAT than the detrended ASI. EOF2 is characterized by an opposite variation be- Comparatively, the correlation between the ASI and the SAT tween Northeast China and the other regions in China is notably lower than the correlation between the BKI and (Figure 4(b)). +e persistent positive value of PC2 sug- PC1. +is implies that the BK is the key region in the Arctic gests a cold anomaly in the western and southern parts of in which the change in the sea ice cover may play a critical China before the late 1990s, resulting in warmer condi- role in the winter SAT anomalies in China. tions in these areas. However, neutral positive 0.4 0.26 0.4 0.4 0.31 0.4 0.31 0.26 0.4 0.26 Advances in Meteorology 7 50°N 50°N 1.2 0.2 40°N 40°N 0.8 0.6 30°N 30°N 20°N 20°N 70°E 80°E 90°E 100°E 110°E 120°E 130°E 140°E 70°E 80°E 90°E 100°E 110°E 120°E 130°E 140°E –1 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1 –1 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1 (a) (b) 2.5 2 2 1.5 1.5 1 1 0.5 0.5 0 0 –0.5 –0.5 –1 –1 –1.5 –1.5 –2 –2 –2.5 –2.5 –3 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 (c) (d) Figure 4: Spatial patterns of the first (a) and the second (b) EOF modes of the winter surface air temperature anomalies at 160 stations in China from 1979 to 2018 (unit: C); the corresponding normalized time coefficients for the first (c) and the second (d) EOF modes. Table 1: Correlation coefficients between the month-to-month detrended (original) BKI, ASI, PC1, and PC2. BKI ASI EOF 9 10 11 9 10 11 ∗ ∗ ∗ PC1 0.48 (0.27) 0.42 (0.18) 0.19 (0.00) 0.40 (0.22) 0.39 (0.29) 0.18 (0.23) PC2 0.03 (− 0.16) 0.18 (0.29) 0.39 (0.47 ) 0.16 (0.33) 0.08 (0.27) 0.19 (0.31) +e bold font and asterisks indicate the significance exceeding 0.05 and 0.01, respectively. correlations are indicated between PC2 and the sea ice 5. Sea Ice Loss-Associated Circulation cover in both the BK and the Arctic except for the BKI in November. To explain how the preceding sea ice variation induces +e above results show a decreasing relationship be- winter SAT anomalies in China, composite differences be- tween the month-to-month variability of the BKI and the tween the sea-level pressure (SLP) and the 850 hPa flow winter SAT in China, which suggests that more attention based on low-high OBKI (DBKI) values in September are should be paid to the preceding September sea ice cover in presented in Figures 5(a) and 5(c). +e difference field shows the BK instead of the signals near October and November, a positive anomalous winter pressure system in the north- which is when we focus on the seasonal prediction of western part of the Eurasian continent that is suggestive of winter SAT anomalies in China. Moreover, the detrended the northwestward expansion and intensification of the Cold BKI time series presents a higher correlation than the Siberian High following the preceding loss of sea ice in the original time series, implying that the interannual vari- BK. Correspondingly, a massive anticyclone in the lower ability in the sea ice cover may play a more critical role in troposphere is present that is superimposed on the high- its association with the SAT in China compared with its pressure system. +is anomalous circulation pattern facili- linear trend component. tates the invasion of cold air from high latitudes to East Asia 8 Advances in Meteorology 80°N 80°N 200 400 70°N 70°N 1 60°N 60°N –1 –2 50°N 50°N –1 –2 –1 40°N 100 40°N 30°N 30°N –100 20°N 20°N 0 15°E 30°E 45°E 60°E 75°E 90°E 105°E 120°E 135°E 150°E 0 15°E 30°E 45°E 60°E 75°E 90°E 105°E 120°E 135°E 150°E (a) (b) 80°N 80°N 2 5 70°N 70°N 60°N 60°N 300 –1 –3 50°N 50°N –1 –2 40°N 40°N –1 –2 30°N 30°N –1 20°N 20°N 0 15°E 30°E 45°E 60°E 75°E 90°E 105°E 120°E 135°E 150°E 0 15°E 30°E 45°E 60°E 75°E 90°E 105°E 120°E 135°E 150°E (c) (d) − 1 Figure 5: Composite differences for the (a) SLP (shaded and contour; unit: Pa), wind (vector; unit: m·s ) at 850 hPa, and (b) surface air temperature (shaded and contour; unit: C) between selected low OBKI years and high OBKI years in September. (c, d) +e same as (a) and (b) but for the composite differences between the selected low DBKI years and high DBKI years in September. +e shading denotes significance exceeding 0.1. +e black vectors indicate winds with a northerly component over the area east of 100 E. and the weakening of warm midlatitude westerlies from the Table 2: Correlation coefficients between the DBKI in September North Atlantic that are instrumental to the cold advection. and the circulation indices in the winter. +is corresponds to large areas of cold anomalies in the midlatitudes of Eurasia as well as East Asia (Figures 5(b) and Circulation indices Correlation coefficient 5(d)). Comparatively, the significant areas of the SAT be- Westerly index 0.42 tween the low-high DBKI in September in Figure 5(d) ex- Eurasian blocking high frequency − 0.65 tend more southeastward to the eastern part of China, the Eurasian anticyclone intensity − 0.55 Korean Peninsula, Japan, and the Northwest Pacific than the Siberian High intensity − 0.45 DBKI in Figure 5(b). A substantial positive high-pressure East Asian winter monsoon index − 0.44 AO index 0.21 system accompanied by a low-pressure system over the Northwest Pacific intensifies the land-sea pressure gradient. +e bold font indicates significance exceeding 0.01. A northerly component correspondingly prevails in the area east of 100 E on the Eurasian continent, which contributes to an intensification of the East Asian winter monsoon. correlate positively with each other at 0.42, indicating fa- However, the northerly component of the low-high OBKI is vorable weakening westerly conditions and a meridional mainly present over the northeastern part of Eurasia. +e circulation pattern following the loss of sea ice in the BK. +e associated response of the intensified East Asian winter significant anticorrelation between the blocking frequency, anticyclone intensity, and Siberian High index and the DBKI monsoon to the low-high DBKI contributes to colder conditions in China compared with the response to the suggest that the retreat of the sea ice cover facilitates the maintenance and intensification of synoptic fluctuations, OBKI in September. +erefore, the following analysis is mainly based on the DBKI in September. such as blocking highs and surface cold highs, which lead to To explore the wintertime atmospheric response over an accumulated strengthening of the semipermanent Sibe- Eurasia to the preceding variation of sea ice, we calculated rian High. Correspondingly, an increase in the land-sea the correlation coefficients between the DBKI in September pressure gradient over East Asia suggests an intensified East and the circulation indices in winter. As shown in Table 2, Asian monsoon that leads to large-scale cold anomalies in most of the circulation indices are closely related to the China. On the contrary, although the AO is well represented September DBKI with significance exceeding 0.01 except for with the zonal mean status of midlatitude westerlies [38], the the AO index. In particular, the DBKI and the westerly index relationship of the westerly index with the DBKI is notably Advances in Meteorology 9 closer to the relationship between the AO and the DBKI. air temperature in the BK shows clearly persistent warming +is implies a local connection of the preceding sea ice in the from September to February and potentially results in a BK with the midlatitude climate anomalies. Kug et al. [14] decrease of the meridional temperature gradient in the also documented different local responses of the variations winter. Coincidently, the autumn heat content of the upper of the sea ice cover in the BK and in the East Siberian and ocean in the BK presents a significantly warm difference Chukchi seas; the variation of the sea ice cover in the BK is between low and high DBKI (Figures 7(a)–7(c)). Moreover, associated well with the winter SAT anomalies over the a slightly warmer condition is induced by low-high Sep- Eurasian continent, while the variation of the sea ice cover in tember DBKI (S-DBKI) compared to the October DBKI (O- the East Siberian and Chukchi seas is mainly linked with the DBKI), and the contrast of oceanic warming is even more SAT anomalies over North America. clear between the S-DBKI and the November DBKI (N- +e reduction in sea ice has a direct connection to the DBKI). +e seasonal sea ice melting from May to September warming of the air column and modifies the meridional opens a large portion of the Arctic Ocean, allowing it to thickness gradient over mid- to high-latitude regions, absorb sunlight during the warm season [40]. +erefore, the leading to the change in midlatitude weather [8, 12]. To retreat of the autumn sea ice cover in the BK associates with examine the linkage between sea ice loss and midlatitude more heat storage in the upper BK, especially in September. circulation, the regional means of the air temperature Correspondingly, the subsequent December-January mean ° ° ° ° profiles over the BK (30 E–100 E, 75 N–85 N, referred to as surface sensible + latent heat flux and upward longwave the north region (NR)) and the midlatitudes of Eurasia radiation show a clear increase in the BK (Figures 7(d)–7(f)). ° ° ° ° (30 E–100 E, 40 N–50 N, referred to as the south region +e positive anomalous ocean-to-atmosphere energy in- (SR)) were calculated, and their composite differences dicates an additional release of oceanic heat storage in the between low-high sea ice years are shown in Figure 6. A BK to the atmosphere associated with the preceding low continuous warming in the NR from September (0 yr) to autumn sea ice cover. +e winter heat anomalies in the BK February (+1 yr) is indicated in the lower troposphere, but and their surroundings are most pronounced to be asso- it decreases with an increase in the height (Figure 6(a)). ciated with the S-DBKI compared to the O-DBKI and Lower-troposphere warming is typically induced by the N-DBKI. release of heat flux from the open ocean to the atmosphere Figure 8 shows the composite differences between the due to the reduction in sea ice [39]. A clear difference in the December-January mean 500 hPa geopotential height warming in the NR is observed between the late autumn (Z500) and the 850 hPa flow (V850) based on low-high and winter. During October and November, the anomalous DBKI values from September to November. A pronounced warming is mainly displayed below 850 hPa with a low high Z500 center is shown in the BK and extends southward significance, and a weak cooling can also be found in the to the Ural Mountains associated with low S-DBKI upper air in November. However, substantial deep (Figure 8(a)). Similar Z500 anomaly patterns also present warming occurs in the midtroposphere and low tropo- through the whole troposphere with a barotropic vertical sphere in the winter with a maximum value of 3.2 C near structure (figure not shown) and link with the amplification the surface. Moreover, dominant cooling is evident in the tropospheric warming induced by sea ice loss. Furthermore, the anomalous southerlies from the high latitudes of SR, especially in January and February (Figure 6(b)). In addition, the September DBKI is associated more closely Northern Atlantic drive warm advection into the BK and facilitate a further amplification warming. +e warming over with midlatitude SAT variations in the winter than in the autumn. Furthermore, the composite differences of di- the BK links with the release of oceanic heat storage to the abetic heating in the NR (SR) from December (0 yr) to atmosphere and associated anomalous warm advection over February (+1 yr) between low-high sea ice years are 39.46 the Barents Sea and cold advection over the midlatitude of − 2 − 2 − 2 (9.04) w·m , 34.14 (− 9.55) w·m , and 29.74(− 7.68) w·m , the Eurasia. A notable difference in temperature between the respectively. +is suggests that the retreat of the sea ice in BK and the midlatitude of Eurasia results in the significant the BK in autumn favors the mid- and lower-tropospheric weakened meridional temperature gradient. +is decelerates warming and thus weakens the meridional temperature the westerlies and leads to more active surface anticyclone between the high latitude and the midlatitude. +e thermal and blocking (Table 2) and anticyclonic vorticity forcing to their north [41]. Correspondingly, large-scale anticyclonic conditions in the BK may play a driving role in the changes in the meridional temperature gradient and the associated circulation is shown over the northwest of Eurasia. As discussed in Section 5, the anomalous anticyclonic circu- thickness gradient. lation over the Ural Mountains and BK leads to more Ural blocking, a deeper East Asian trough, and more intense cold 6. Month-to-Month Oceanic Heat Storage air invasion into the East Asian region. In contrast, the low O-DBKI and N-DBKI associated Z500 anomalies display a Contrast and Associated northwestward-displaced positive center over the west of Subsequent Warming Greenland (Figures 8(b) and 8(c)). Cold advection from the In this section, we will address the reason why the DBKI in east Siberian sector of Arctic is weaker compared to S-DBKI- September could be a better indicator of the SAT anomalies associated Z500 anomalies (Figure 8(a)). Consequently, the S-DBKI shows a higher correlation with winter SAT in China compared to the DBKI in the following October and November. As discussed above, the lower-troposphere anomalies in China. 10 Advances in Meteorology hPa hPa hPa 500 500 500 600 600 600 700 700 700 775 775 775 850 850 850 0.8 925 0.8 925 925 1.6 –0.8 1000 1000 1000 Sep Oct Nov Dec Jan Feb Sep Oct Nov Dec Jan Feb Sep Oct Nov Dec Jan Feb (a) (b) (c) ° ° ° ° ° Figure 6: Composite differences for (a) vertical profiles of the temperature (shaded; unit: C) in the BK region/NR (30 E–100 E, 75 N–85 N) from September to the following February between selected low DBKI years and high DBKI years in September. (b, c) +e same as (a) but for ° ° ° ° the midlatitude region/SR (30 E–100 E, 40 N–50 N) and the differences between the SR and the NR, respectively. +e green dot denotes the values that are statistically significant with significance exceeding 0.1 based on the results of t-tests. 90°W 90°W 90°W 60°W 120°W 60°W 120°W 60°W 120°W 30°W 150°W 150°W 30°W 150°W 30°W 0° 180° 0° 180° 0° 180° 150°E 150°E 150°E 30°E 30°E 30°E 120°E 120°E 60°E 60°E 60°E 120°E 90°E 90°E 90°E Contour from –0.8 to 1.4 by 0.244 Contour from –0.8 to 1.4 by 0.244 Contour from –0.8 to 1.4 by 0.244 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1.1 1.4 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1.1 1.4 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1.1 1.4 (a) (b) (c) 90°W 90°W 90°W 60°W 120°W 60°W 120°W 60°W 120°W 150°W 150°W 150°W 30°W 30°W 30°W 0° 180° 0° 180° 0° 180° 30°E 150°E 30E 150°E 30°E 150°E 60°E 120°E 60°E 120°E 60°E 120°E 90°E 90°E 90°E Contour from –100 to 100 by 20 Contour from –100 to 80 by 20 Contour from –140 to 140 by 20 –20 –10 10 20 30 40 50 60 –20 –10 10 20 30 40 50 60 –20 –10 10 20 30 40 50 60 (d) (e) (f ) Figure 7: Composite differences for the SON mean upper ocean heat content (shaded) between selected low and high DBKI in September (a), October (b), and November (c) (unit: K). (d–f) +e same as (a), (b), and (c) but for DJF mean upward longwave radial (shaded) and − 2 sensible + latent heat flux (contour) (unit: w·m ). +e dotted areas denote the values that are statistically significant with significance exceeding 0.1 based on the results of t-tests for heat content in (a), (b), and (c) and longwave radiation in (d), (e), and (f). –4.8 –4 –3.2 –2.4 –1.6 –0.8 0.8 1.6 2.4 3.2 4.8 –4.8 –4 –3.2 –2.4 –1.6 –0.8 0.8 1.6 2.4 3.2 4.8 –4.8 –4 –3.2 –2.4 –1.6 –0.8 0.8 1.6 2.4 3.2 4.8 Advances in Meteorology 11 90°W 90°W 90°W 60°W 120°W 60°W 120°W 60°W 120°W 150°W 150°W 30°W 30°W 30°W 150°W 0° 0° 0° 180° 180° 180° 150°E 150°E 30°E 30°E 150°E 30°E 6 6 120°E 60°E 120°E 60°E 60°E 120°E 90°E 90°E 90°E –80 –60 –40 –20 –10 10 20 40 60 80 –80 –60 –40 –20 –10 10 20 40 60 80 –80 –60 –40 –20 –10 10 20 40 60 80 (a) (b) (c) − 1 Figure 8: +e same as (d), (e), and (f) in Figure 7 but for DJF mean Z500 (shaded) (unit: gpm) and V850 (vector) (unit: m·s ). +e dotted areas denote the values that are statistically significant with significance exceeding 0.1 based on the results of t-tests for Z500. +is regional tropospheric warming results in a 7. Conclusions higher barotropic positive height anomaly over the +e variation of the autumn Arctic sea ice area and the re- Ural Mountains, and thus, more active cold advec- lationship between the month-to-month sea ice and the winter tion from the high latitude affects East Asia. temperature anomalies in China are investigated in this study. +e conclusions based on the main results are as follows: Data Availability (1) +e sea ice cover in the BK during the autumn season +e data used to support the findings of this study are in- has shown a substantial decrease during the past cluded within the article: (1) the European Centre for three decades. +e region of accelerating sea ice Medium-Range Weather Forecasts (ECMWF) reanalysis reduction since 2005 agrees well with the Arctic dataset (ERA-Interim) from 1979 to 2018 is used for deriving superimposed with a large interannual variability. data on the daily and monthly air temperatures, sea-level (2) +e retreat of autumn sea ice in the BK is significantly pressure, geopotential height, and horizontal wind; (2) associated with large-scale negative temperature monthly mean sea ice concentrations are derived from the anomalies in the following winter in China, and this Met Office Hadley Centre’s sea ice and sea surface tem- retreat correlates well with the dominant sign-con- perature (HadiSST 1.1) dataset; (3) the monthly mean heat sistent SAT mode (EOF1) in China, except for the content is derived from the vertical average of potential Tibetan Plateau. However, the ice-SAT lag correlation temperature (109.8 m above; 11 levels) from the ECMWF shows a notable month-to-month diversity in which ocean reanalysis system ORAS4; and (4) observed monthly the linkage between the sea ice in the BK in September mean surface air temperatures (SATs) are from the National and the winter temperature is stronger than that in Climate Center of China, including 160 stations. both October and November. Moreover, the detrended sea ice cover in the BK as a potential Conflicts of Interest predictor is associated more closely with the SAT in China than the original sea ice cover in the BK. +e authors declare that there are no conflicts of interest (3) An anomalous positive surface pressure is exhibited regarding the publication of this paper. over the northwestern part of Eurasia in the winter that is linked to the decreasing sea ice in the BK region Acknowledgments in the preceding September. +is surface pressure favors the persistence and intensification of synoptic +is work was supported jointly by the National Basic perturbations, such as blocking highs and surface cold Research Program of China (Grant no. 2015CB953904) and highs, as well as the intensification of the Siberian the National Natural Science Foundation of China (Grant High and East Asian winter monsoon. +ese favorable nos. 41975073, 41575081, and 41741005). conditions ultimately contribute to the formation of large-scale winter cold anomalies in China. References (4) Compared to low O-DBKI and N-DBKI, a more oceanic heat storage in the upper BK is induced by [1] T. R. Karl, A. Arguez, B. Huang et al., “Possible artifacts of low S-DBKI leading to a larger heat release to tro- data biases in the recent global surface warming hiatus,” pospheric atmosphere by surface heat flux in the BK. Science, vol. 348, no. 6242, pp. 1469–1472, 2015. 12 Advances in Meteorology [2] X. Zhang, C. Lu, and Z. Guan, “Weakened cyclones, in- Meteorologica Sinica, vol. 16, no. 4, pp. 34–40, 1994, in tensified anticyclones and recent extreme cold winter weather Chinese. events in Eurasia,” Environmental Research Letters, vol. 7, [20] X. M. Shi, Y. W. Sun, and J. L. Sun, “+e impact of zonal no. 4, Article ID 044044, 2012. temperature gradient in the Arctic Ocean on the summer [3] C. H. Lu, S. Q. Xie, Y. J. Qin et al., “Recent intensified winter precipitation over Northeast China,” Periodical of Ocean coldness in the mid-high latitudes of Eurasia and its re- University of China, vol. 44, no. 2, pp. 11–16, 2014, in Chinese. lationship with daily extreme low temperature variability,” [21] H. J. Wang, H. P. Chen, and J. Liu, “Arctic sea ice decline Advances in Meteorology, vol. 2016, Article ID 3679291, intensified haze pollution in Eastern China,” Atmospheric and 11 pages, 2016. Oceanic Science Letters, vol. 8, no. 1, pp. 1–9, 2015. [4] L. Yang and B. Wu, “Interdecadal variations of the East Asian [22] B. Y. Wu, R. H. Huang, and D. Y. Gao, “+e impact of variation winter surface air temperature and possible causes,” Chinese of sea-ice extent in the Kara Sea and the Barents Seas in winter on Science Bulletin, vol. 58, no. 32, pp. 3969–3977, 2013. the winter monsoon over East Asia,” Chinese Journal of At- [5] L. Wang and W. Chen, “+e East Asian winter monsoon: Re- mospheric Sciences, vol. 23, no. 3, pp. 267–275, 1999, in Chinese. amplification in the mid-2000s,” Chinese Science Bulletin, [23] X. H. Xie and X. Q. Yang, “Interannual relationship between vol. 59, no. 4, pp. 430–436, 2014. Arctic sea ice concentration anomaly and China air tem- [6] J. A. Francis, “Why are Arctic linkages to extreme weather still perature variation during winter,” Journal of Nanjing Uni- up in the air?,” Bulletin of the American Meteorological Society, versity (Natual Sciences), vol. 42, no. 6, pp. 549–561, 2006. vol. 98, no. 12, pp. 2551–2557, 2017. [24] B. Y. Wu, J. Z. Su, and R. H. Zhang, “Effects of autumn-winter [7] J. Overland, “Potential arctic change through climate am- Arctic sea ice on winter Siberian High,” Chinese Science plification processes,” Oceanography, vol. 24, no. 3, pp. 176– Bulletin, vol. 56, no. 30, pp. 3220–3228, 2011. 185, 2011. [25] F. Li, H. J. Wang, and Y. Q. Gao, “Change in sea ice cover is [8] J. A. Francis and S. J. Vavrus, “Evidence linking Arctic am- responsible for non-uniform variation in winter temperature plification to extreme weather in mid-latitudes,” Geophysical over East Asia,” Atmospheric and Oceanic Science Letters, Research Letters, vol. 39, no. 6, Article ID L06801, 2012. vol. 8, no. 6, pp. 376–382, 2015. [9] B.-M. Kim, S.-W. Son, S.-K. Min et al., “Weakening of the [26] C. Y. Zhu, F. Huang, Y. H. Shi et al., “Spatial-temporal stratospheric polar vortex by Arctic sea-ice loss,” Nature patterns of the cold surge events in China in recent 50 years Communications, vol. 5, no. 1, 2014. and its relationship with Arctic sea ice,” Periodical of Ocean [10] E. A. Barnes, “Revisiting the evidence linking Arctic ampli- University of China, vol. 44, no. 12, pp. 12–20, 2014, in fication to extreme weather in midlatitudes,” Geophysical Chinese. Research Letters, vol. 40, no. 17, pp. 4734–4739, 2013. [27] C. Li and Q. Zhang, “January temperature anomalies over [11] C. Deser and H. Teng, “Evolution of Arctic sea ice concen- Northeast China and precursors,” Chinese Science Bulletin, tration trends and the role of atmospheric circulation forcing vol. 58, no. 6, pp. 671–677, 2013. 1979–2007,” Geophysical Research Letters, vol. 35, no. 2, [28] D. W. Wang and X. Q. Yang, “Temporal and spatial patterns Article ID L02504, 2008. of Arctic sea ice variations,” Acta Meteorologica Sinica, vol. 60, [12] J. E. Overland and M. Wang, “Large-scale atmospheric cir- no. 2, pp. 129–138, 2002. culation changes are associated with the recent loss of Arctic [29] J. H. He, F. M. Wu, and L. Qi, “Decadal/interannual linking sea ice,” Tellus A: Dynamic Meteorology and Oceanography, between autumn Arctic sea ice and following winter Eurasian vol. 62, no. 1, pp. 1–9, 2010. air temperature,” Chinese Journal of Geophysics, vol. 58, no. 4, [13] M. Mori, M. Watanabe, H. Shiogama, J. Inoue, and pp. 1089–1102, 2015, in Chinese. M. Kimoto, “Robust Arctic sea-ice influence on the frequent [30] D. P. Dee, S. M. Uppala, A. J. Simmons et al., “+e ERA- Eurasian cold winters in past decades,” Nature Geoscience, Interim reanalysis: configuration and performance of the data vol. 7, no. 12, pp. 869–873, 2014. assimilation system,” Quarterly Journal of the Royal Meteo- [14] J. S. Kug, J. H. Jeong, Y. S. Jang et al., “Two distinct influences rological Society, vol. 137, no. 656, pp. 553–597, 2011. of Arctic warming on cold winters over North America and [31] N. A. Rayner, D. E. Parker, and E. B. Horton, “Global analyses East Asia,” Nature Geoscience, vol. 8, no. 10, Article ID of sea surface temperature, sea ice, and night marine air NGEO2517, pp. 759-762, 2015. temperature since the late nineteenth century,” Journal of [15] P. Ruggieri, R. Buizza, and G. Visconti, “On the link between Geophysical Research, vol. 108, no. D14, 2003. Barents-Kara sea ice variability and European blocking,” [32] M. A. Balmaseda, K. Mogensen, and A. T. Weaver, “Evalu- Journal of Geophysical Research: Atmospheres, vol. 121, no. 10, ation of the ECMWF ocean reanalysis system ORAS4,” pp. 5665–5679, 2015. Quarterly Journal of the Royal Meteorological Society, vol. 139, [16] Z. Lu, ¨ S. He, F. Li, and H. Wang, “Impacts of the autumn no. 674, pp. 1132–1161, 2013. Arctic sea ice on the intraseasonal reversal of the winter [33] D. Y. Gong and C. H. Ho, “+e Siberian High and climate Siberian High,” Advances in Atmospheric Sciences, vol. 36, change over middle to high latitude Asia,” 9eoretical and no. 2, pp. 173–188, 2019. Applied Climatology, vol. 72, no. 1-2, pp. 1–9, 2002. [17] Q. Xie and S. S. Huang, “Study of influence of equatorial SST [34] J. G. Charney, J. Shukla, and K. C. Mo, “Comparison of a over Mid-East Pacific ocean and Arctic sea-ice anomalies to barotropic blocking theory with observation,” Journal of the general circulation,” Scientia Meteorologica Sinica, vol. 10, Atmospheric Sciences, vol. 38, no. 4, pp. 762–779, 1981. no. 4, pp. 325–338, 1990. [35] N. Shi and Y. S. Yang, “Main characteristics of east Asian [18] B. Wu, D. Gao, and R. Huang, “ENSO events and interannual summer/winter intermonsoon index for 1873∼1996,” Journal variations of winter sea-ice in the Greenland, the Kara and the Barents Seas,” Chinese Science Bulletin, vol. 42, no. 16, of Nanjing Institute of Meteorology, vol. 21, no. 2, pp. 208–214, pp. 1382–1385, 1997. 1998, in Chinese. [19] X. Q. Yang, Q. Xie, and S. S. Huang, “Modelling influence of [36] J. Inoue and E. H. Masatake, “+e role of Barents sea ice in the Arctic sea-ice anomalies to Asian summer monsoon,” Acta wintertime cyclone track and emergence of a warm-Arctic Advances in Meteorology 13 cold-Siberian anomaly,” Journal of Climate, vol. 25, no. 7, [37] G. R. North, T. L. Bell, R. F. Cahalan, and F. J. Moeng, “Sampling errors in the estimation of empirical orthogonal functions,” Monthly Weather Review, vol. 110, no. 7, pp. 699–706, 1982. [38] D. W. J. +ompson and J. M. Wallace, “Annular modes in the extratropical circulation. Part I: month-to-month variability,” Journal of Climate, vol. 13, no. 5, pp. 1000–1016, 2000. [39] J. A. Screen and I. Simmonds, “+e central role of diminishing sea ice in recent Arctic temperature amplification,” Nature, vol. 464, no. 7293, pp. 1334–1337, 2010. [40] A. Dai, D. Luo, M. Song et al., “Amplification is caused by sea- ice loss under increasing CO ,” Nature Communications, vol. 10, no. 121, pp. 1–13, 2019. [41] N.-C. Lau, “Variability of the observed midlatitude storm tracks in relation to low-frequency changes in the circulation pattern,” Journal of the Atmospheric Sciences, vol. 45, no. 19, pp. 2718–2743, 1988. International Journal of The Scientific Advances in Advances in Geophysics Chemistry Scientica World Journal Public Health Hindawi Hindawi Hindawi Hindawi Publishing Corporation Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 Journal of Environmental and Public Health Advances in Meteorology Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Applied & Environmental Journal of Soil Science Geological Research Hindawi Volume 2018 Hindawi www.hindawi.com www.hindawi.com Volume 2018 International Journal of International Journal of Agronomy Ecology International Journal of Advances in International Journal of Forestry Research Microbiology Agriculture Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 International Journal of Journal of Journal of International Journal of Biodiversity Archaea Analytical Chemistry Chemistry Marine Biology Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Advances in Meteorology Hindawi Publishing Corporation

Month-to-Month Variability of Autumn Sea Ice in the Barents and Kara Seas and Its Relationship to Winter Air Temperature in China

Download PDF
14 pages

Loading next page...
 
/lp/hindawi-publishing-corporation/month-to-month-variability-of-autumn-sea-ice-in-the-barents-and-kara-9XBiXw65ZY

References (44)

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2019 Chuhan Lu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
1687-9309
eISSN
1687-9317
DOI
10.1155/2019/4381438
Publisher site
See Article on Publisher Site

Abstract

Hindawi Advances in Meteorology Volume 2019, Article ID 4381438, 13 pages https://doi.org/10.1155/2019/4381438 Research Article Month-to-Month Variability of Autumn Sea Ice in the Barents and Kara Seas and Its Relationship to Winter Air Temperature in China 1 1 2 3 1 Chuhan Lu , Kaili Li, Shaoqing Xie, Zhaomin Wang, and Yujing Qin Key Laboratory of Meteorological Disaster, Ministry of Education, Joint International Research Laboratory of Climate and Environment Change, Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing 210044, China Jiashan Meteorological Bureau, Jiaxing 314100, China College of Oceanography, Hohai University, Nanjing 210000, China Correspondence should be addressed to Yujing Qin; qinyujing@nuist.edu.cn Received 27 May 2019; Revised 30 August 2019; Accepted 29 October 2019; Published 12 December 2019 Academic Editor: Alastair Williams Copyright © 2019 Chuhan Lu et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +e variation of autumn Arctic sea ice is a critical indicator of temperature anomalies over the Eurasian continent during winter. +e retreat of autumn Arctic sea ice is typically accompanied by negative anomalous winter temperatures over the Eurasian and North American continents. However, such sea ice temperature linkages notably change from month to month. +e variation of the autumn Arctic sea ice area and the relationship between the month-to-month sea ice and winter temperature anomalies in China are investigated using the Hadley Centre’s sea ice dataset (HadiSST) and the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis dataset (ERA-Interim) during 1979–2018. We present the following results: +e sea ice in the Barents and Kara seas (BK) during the autumn and winter seasons shows notable low-frequency variability. +e retreat of sea ice in the BK from September to November is significantly associated with negative temperature anomalies in the following winter in China. However, the linkage between the sea ice in the BK in September and the winter temperatures is stronger than that in both October and November. An anomalous positive surface pressure is exhibited over the northwestern part of Eurasia in the winter that is linked to decreasing sea ice in the BK in the preceding September. +is surface pressure favors the persistence and intensification of synoptic perturbations, such as blocking highs and surface cold highs, as well as the intensification of the Siberian High and the East Asian winter monsoon. +ese favorable conditions ultimately contribute to the formation of large-scale winter cold anomalies in China. Compared to low sea ice cover in October and November, a more oceanic heat storage in the upper BK induced by low sea ice cover in the BK leads to a larger heat release to tropospheric atmosphere in winter by surface heat flux and upward longwave radiation in the BK. +is regional tropospheric warming results in a higher barotropic positive height anomaly over the Ural Mountains, and then more active cold advection from the high latitude affects East Asia. [2, 3]. Along with this unforeseen cooling trend in the 1. Introduction midlatitude regions of the NH, both the Cold Siberian High Arctic sea ice is a prominent indicator of global climate and the East Asian winter monsoon exhibit a substantial change. A substantial loss of ice and a higher warming rate decadal intensification, leading to more active cold air have been observed in the Arctic during recent decades, a outbreaks in China [4, 5]. Recent studies documented that phenomenon known as Arctic amplification, with a con- there may be some physical linkages between the recent tinuous increase in the global air temperature [1]. Severe reduction of ice in the Arctic Ocean and cooling in the cold events and anomalously heavy snowfalls have fre- midlatitude continents of the NH through the anomalous quently affected most midlatitude regions of the Northern atmospheric circulation response of sea ice loss [6]. For Hemisphere (NH), including major industrialized centers example, a reduction in the Arctic sea ice is typically 2 Advances in Meteorology accompanied by substantial warming that weakens the reduction that occurs east of the Arctic Ocean and the BK, is meridional temperature gradient. +is will decelerate the associated with substantial surface warming. A weakening westerlies in the midlatitudes and result in a decrease in polar/midlatitude temperature gradient results in a Eurasian atmospheric baroclinicity and a more wavy circulation [7, 8]. (EU) pattern response that facilitates the strengthening of Compared with the direct response within the troposphere, the Siberian High and the East Asian winter monsoon, stratospheric response to sea ice forcing also plays a robust which brings substantial cooling and more extreme low role in the development of cold conditions over midlatitude events in East Asia [22–24]. Li et al. [25] documented the continents of the NH [9]. +ere is still a debate regarding the local responses of the temperature in East Asia to changes in physical connection between the changes in Arctic sea ice sea ice in different regions. +ey reported that the retreat of and the midlatitude activity of planetary waves [10]. A the sea ice in the BK induces an intensification of the change in the dominant mode of extratropical circulation Siberian High and lowers the temperature in the area north may also have a close association with the recent cooling of East Asia, while an increase in sea ice in the Okhotsk Sea trend [11–13]. induces a northerly shift of the jet stream in the upper Previous studies have shown that extratropical impacts troposphere over East Asia with substantial warming in the depend highly on the regional structure of the anomalous area south of East Asia. +e variations of the sea ice in the BK Arctic climate state [14]. For example, warming over the BK are associated with the Arctic Oscillation (Arctic Dipole region can lead to East Asian cooling, whereas northern pattern) by inducing cross-Arctic-like (Eurasian) telecon- North America cooling is closely related to warming over the nection wave-train patterns that lead to an intensification of East Siberian Sea–Chukchi Sea region [14]. +e reduction of cold surge activity in China [26]. +e SST anomalies in the sea ice over the BK also has a cooling effect on the European NH and the changes in Arctic sea ice were involved in the region induced by more blockings associated with the loss of interdecadal changes in the winter surface air temperature sea ice in the BK [15]. However, atmospheric responses to (SAT) over East Asia that occurred approximately during the sea ice variations of the BK vary with background at- the mid-1990s [4, 27]. mospheric states and seasons. A recent study had pointed It is worthy to note that the variation of the sea ice cover shows clear seasonal and regional dependencies [28], in out that the loss of sea ice in the BK in November links with a weaker SH in November, while a stronger SH in December spite of a general decrease trend of the Arctic sea ice. [16]. Furthermore, the relationships between wintertime SAT Many previous studies have focused on the relationship anomalies in China and sea ice variation in different time between the variations in Arctic sea ice and atmospheric scales or regions also display notable diversity [23, 29]. circulation patterns and the associated anomalous climate in +erefore, we will study the association of month-to- China. Xie and Huang [17] reported that changes in Arctic month variability of the regional Arctic sea ice cover in the sea ice and sea surface temperature (SST) anomalies in the preceding autumn with the SAT anomalies and its possible central and eastern equatorial Pacific play important roles in linking pathway. the general circulation and suggested that there may be physical connections between them. Wu et al. [18] con- 2. Data and Methods firmed that the variations in the areas of sea ice in the Greenland Sea and BK are associated with the occurrence of 2.1. Data. +e data used in this study include the following: El Niño-Southern Oscillation (ENSO) events and that a 3- (1) the European Centre for Medium-Range Weather year leading signal of anomalous sea ice shows a close Forecasts (ECMWF) reanalysis dataset (ERA-Interim) from linkage with the general circulation in the NH. A numerical 1979 to 2018 is used for deriving data on the daily and simulation also indicated that the variation in Arctic sea ice monthly air temperatures, sea-level pressure, geopotential is one of the most important indicators of an East Asian ° ° height, and horizontal wind with a 2.5 × 2.5 horizontal summer monsoon climate anomaly [19]. Shi et al. [20] resolution [30]; (2) monthly mean sea ice concentrations are documented that there is a positive correlation between the derived from the Met Office Hadley Centre’s sea ice and sea zonal temperature gradient of the Arctic Ocean and the surface temperature (HadiSST 1.1) dataset [31]; (3) the geopotential height gradient between the Barents Sea and the monthly mean heat content is derived from the vertical East Siberian Sea at 500 hPa and that the associated average of potential temperature (109.8 m above; 11 levels) anomalous circulation links with the summer precipitation from the ECMWF ocean reanalysis system ORAS4 [32]; and anomalies over Northeast China. Wang et al. [21] also re- (4) observed monthly mean surface air temperatures (SATs) ported that the preceding autumn loss of Arctic sea ice favors are from the National Climate Center of China, including a northward shift of the cyclone track in China along with 160 stations. the weakening of Rossby waves over the area south of 40 N in eastern China, resulting in an intensification of haze weather. 2.2. Circulation Indices. +e definitions of circulation in- +e influences of changes in Arctic sea ice on the dices are as follows: +e westerly index is defined as the wintertime air temperature in China under the background differences in the geopotential heights between 40 N and ° ° ° of Arctic amplification have drawn extensive attention from 65 N and between 0 and 122 E [33]. +e blocking high scholars in recent years. A series of studies showed that the frequency is derived from the anomalous 500 hPa mid- preceding reduction of Arctic sea ice in autumn, such as the latitude geopotential height [34]. +e anticyclone intensity Advances in Meteorology 3 over Eurasia is the seasonal mean of the center pressure for simultaneous correlation coefficients of the OBKI with the each surface high in 6 hourly SLP field during the winter [2]. Arctic sea ice area index (ASI (downloaded from NSIDC ° ° ° ° +e regional mean of the SLP (40 N–60 N, 70 E–120 E) (http://nsidc.org/data/g02135.html))) in September, Oc- represents the intensity of the Cold Siberian High [33], and tober, and November are 0.43, 0.55, and 0.47, respectively, the East Asian winter monsoon index is defined by the which are statistically significant with significance ex- midlatitude land-sea pressure differences between 110 E and ceeding 0.05. +ese significant correlations are suggestive 160 E [35]. In addition, the monthly Arctic Oscillation (AO) of the consistency of the low-frequency sea ice variations index (http://www.cpc.ncep.noaa.gov/products/precip/ between our defined BK region and the Arctic. +e DBKI CWlink/daily_ao_index/ao.shtml) can be downloaded time series shows distinct differences from the OBKI time from the NOAA Climate Prediction Center (CPC). series during heavy and light ice cover years (defined as ± 0.75 σ), as indicated in Figures 2(d)–2(f ). Additional heavy (light) ice cover years occur before (after) the mid- 3. Variability and Linear Trend of the Arctic 1990s in the OBKI time series that bear an important Sea Ice decreasing feature. A further discussion of different as- To show the horizontal distribution of the Arctic sea ice, sociations between the OBKI and the DBKI is presented in the following section. the standard deviation (σ) of the sea ice concentration data from September to December from 1979 to 2018 is shown in Figure 1. Large value areas generally extend from 4. Relationship between the Sea Ice and the the interior Arctic to the high-latitude oceans in the NH Air Temperature and suggest large interannual variability of the sea ice cover in the marginal areas of the Arctic sea ice. However, Both the OBKI and the DBKI exhibit positive correlations clear local characteristics are shown in the above large with the wintertime SAT anomalies in most parts of China, values of σ areas. In particular, large variability of sea ice as shown in Figure 3, despite their inconsistent time evo- ° ° occurs in the 70 N–80 N region of the Barents, Kara, and lutions. As documented in Figures 3(a)–3(c), there are Beaufort seas in September and October (Figures 1 and significant positive correlations between the OBKI and the 1(b)). +e northwest coast of Canada and the east coast of SAT in North China. +is is indicated by a cold anomaly Greenland also display high σ values. As a continuous likely followed by a preceding autumn sea ice loss in the BK, southward expansion of the ice cover, the areas of high σ which is in agreement with the previous studies by Wu et al. shift from the preceding Laptev and Beaufort seas to the [22, 24]. However, the extent of the association decreases Bering Sea and the northeast coast of Canada after No- from September to November. In particular, significant vember (Figures 1(c) and 1(d)). It should be noted that correlations can be observed over Xinjiang and Northeast there is a continuous large variability in the BK from China; the largest area is observed in September, after which September to December. Since the change in sea ice results it shrinks gradually in October and November. +e com- from strong sea ice-atmosphere coupling, high σ values in posite differences of the SAT between the low and high the BK indicate that the interannual/interdecadal varia- OBKI years also show a correlated consistency. In Sep- tion of sea ice in this region may be related to notable tember, low sea ice likely induces a decrease in the SAT in changes in atmospheric and oceanic interactions under North China, with maximum decreases occurring especially the background of Arctic amplification [36]. In addition, below − 1.6 C. In addition, slight cooling is also evident in the variation of the sea ice in the BK can be a good in- Southwest China. +e correlation patterns in October and dicator of the East Asian winter monsoon [24]. To un- November are similar to the September one with a slight derstand the effect of sea ice variability on the winter air reduction of the number of stations with statistical temperature variability in China, we used the time series significance. of the sea ice cover area over the BK (box area in Figure 1) Stronger associations of the DBKI with the SAT in China as the sea ice index (BKI) in this key region. in the autumn are observed in Figures 3(d)–3(f), and the +e original and detrended time series of the BKI, month-to-month correlation variations also decrease similar hereafter referred to as the OBKI and DBKI, respectively, to the association of the OBKI with the SAT. In September, during September-October-November (SON) from 1979 to pronounced areas of larger significance are observed over 2018 are presented in Figure 2. +e pronounced retreat of most of China compared with the OBKI (73.8% stations with autumn sea ice in the BK during the past 40 yrs is indicated significance exceeding 0.05), except for the Tibetan Plateau by the OBKI in Figures 1(a)–1(c) with values of and its vicinity (Figure 3(d)). +e composite differences in 5 2 − 1 5 2 − 1 − 0.81 × 10 the SAT between low and high DBKI years are generally km ·10 yr (September), − 1.36 ×10 km ·10 yr 5 2 − 1 (October), and − 1.19 ×10 km ·10 yr (November), which − 0.8 C in the significant regions. +e significant results are all statistically significant exceeding 0.01. +e different described above represent a good indication of the variation decreasing trends of the BKI in the autumn may be because of the sea ice in the BK in September when compared with the Arctic sea ice areas in different months correspond to the SAT in China in the following winter. In October, dissimilar general circulations [34]. +e remarkable retreat significant SAT anomaly regions are observed in Xinjiang, of autumn sea ice in the BK is evident after 2005, which is Northeast China, North China, and the Changjiang-Huaihe consistent with the accelerated decrease in the area of Arctic River Basin, which decreases in size in November similar to sea ice over the last decade [35]. In addition, the Northeast China. 4 Advances in Meteorology 90°W 90°W 60°W 120°W 60°W 120°W 150°W 30°W 150°W 30°W 0° 180° 0° 180° 150°E 150°E 30°E 30°E 60°E 120°E 60°E 120°E 90°E 90°E 0.1 0.2 0.3 0.4 0.1 0.2 0.3 0.4 (a) (b) 90°W 90°W 60°W 120°W 60°W 120°W 30°W 30°W 150°W 150°W 0° 180° 0° 180° 0.1 150°E 150°E 30°E 30°E 120°E 60°E 120°E 60°E 90°E 90°E 0.1 0.2 0.3 0.4 0.1 0.2 0.3 0.4 (c) (d) Figure 1: Standard deviations of the Arctic sea ice concentration data for September (a), October (b), November (c), and December (d) from 1979 to 2018. +e black dashed box denotes the Barents Sea and the Kara Sea region. To further study the relationship between the sea ice and time series (PC1) are shown in Figures 4(a) and 4(c). +e the SAT, we conducted an empirical orthogonal function EOF1 mode shows a pattern of consistent positive SAT (EOF) analysis for the station SAT anomalies in the winter. departures in most of China (except the Tibetan Plateau) +e variance contribution rates for the first two modes are with a general north-to-south decrease in SAT values. +is 50.7% and 19.3%, which can be well defined according to the indicates either consistent cooling or consistent warming of criterion proposed by North et al. [37]. +e spatial distri- China in the winter. A comprehensive analysis using EOF1 bution of the leading mode (EOF1) and its corresponding and PC1 indicates that a dominant cooling trend emerged in Advances in Meteorology 5 0.8 1.1 0.7 0.6 0.9 0.8 0.5 0.7 0.4 0.6 0.5 0.3 0.4 0.2 0.3 0.2 0.1 0.1 0 0 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 (a) (b) 2.5 1.3 1.2 1.5 1.1 0.5 0.9 0.8 0 0.7 –0.5 0.6 –1 0.5 –1.5 0.4 –2 0.3 0.2 –2.5 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 (c) (d) 2.5 2 1.5 1.5 0.5 –0.5 0.5 –1 –1.5 –0.5 –2 –2.5 –1 –3 –1.5 –3.5 –2 –4 –2.5 –4.5 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 (e) (f) Figure 2: BKI time series for September (a, d), October (b, e), and November (c, f) showing the time series of both the OBKI in the left 5 2 panels (unit: 10 km ) and the DBKI in the right panels. +e straight lines in the left panels denote the linear trends in the OBKI. +ose years with high BKI values (>0.75 SD) and low BKI values (<− 0.75 SD) are marked by circles and black dots, respectively. the mid-1980s and after 2010, while a prominent warming positive correlations between the OBKI, DBKI, and PC1 are trend was dominant in the late 1980s and 2000s with a larger evident, and the correlation between the DBKI and PC1 is SAT departure amplitude in the northern part of China. In clearly higher than that between the OBKI and PC1. A addition, a notable interannual fluctuation can be observed notable correlation decrease is evident from September to since the late 1990s which is confirmed by a wavelet analysis November. Particularly, the correlation coefficient between the DBKI in September and PC1 is 0.48 with significance that shows a substantial 2- to 4-year fluctuation since the early 21st century (figure not shown). exceeding 0.01, while the correlation between the DBKI and To further assess the correlation between the SAT PC1 in November decreases to 0.19. +e significant corre- principal component and the sea ice variation, the corre- lations between the sea ice in the BK and PC1 further suggest lation coefficients between the month-to-month OBKI, that the change in the sea ice in the BK in the autumn can be DBKI, ASI, and PC1 and PC2 were calculated, the results of utilized as a skillful predictor of the winter SAT anomalies in which are documented in Table 1. +e difference between the China, especially for the DBKI in September. However, month-to-month OBKI and DBKI is statistically significant similar decreases in the correlations between the ASI and with significance exceeding 0.01. Moreover, significant PC1 are evident, and the original ASI also shows a lower 0.31 0.26 0.26 0.26 0.31 6 Advances in Meteorology 50°N 50°N 0.05 0.05 0.1 0.26 0.1 0.26 40°N 40°N 0.31 0.31 0 0 0.26 0 0.26 0.31 –0.26 –0.1 –0.1 0.31 –0.4 30°N 30°N –0.05 –0.05 0.26 –0.01 –0.01 20°N 20°N 80°E 100°E 120°E 140°E 80°E 100°E 120°E 140°E –0.3~–0.8 –1.6~–2.4 –0.3~–0.8 –1.6~–2.4 0.3~0.8 1.6~2.4 0.3~0.8 1.6~2.4 –0.8~–1.6 <–2.4 –0.8~–1.6 <–2.4 0.8~1.6 >2.4 0.8~1.6 >2.4 (a) (b) 50°N 50°N 0.05 0.05 0.26 0.1 0.1 0.31 40°N 40°N 0.4 0.4 0 0.31 0 0.26 0.31 0.31 0.26 –0.1 –0.1 0.4 30°N 30°N –0.05 –0.05 –0.01 –0.01 20°N 20°N 80°E 100°E 120°E 140°E 80°E 100°E 120°E 140°E –0.3~–0.8 –1.6~–2.4 –0.3~–0.8 –1.6~–2.4 0.3~0.8 1.6~2.4 0.3~0.8 1.6~2.4 –0.8~–1.6 <–2.4 –0.8~–1.6 <–2.4 0.8~1.6 >2.4 0.8~1.6 >2.4 (c) (d) 50°N 0.05 50°N 0.05 0.26 0.31 0.31 0.1 0.1 40°N 0.31 40°N 0.31 0.26 0 0 0.26 0.31 –0.1 –0.1 30°N 30°N –0.05 –0.05 0.26 –0.01 –0.01 20°N 20°N 80°E 100°E 120°E 140°E 80°E 100°E 120°E 140°E –0.3~–0.8 –1.6~–2.4 –0.3~–0.8 –1.6~–2.4 0.3~0.8 1.6~2.4 0.3~0.8 1.6~2.4 –0.8~–1.6 <–2.4 –0.8~–1.6 <–2.4 0.8~1.6 >2.4 0.8~1.6 >2.4 (e) (f) Figure 3: Correlation coefficients between the BKI for September (a, d), October (b, e), and November (c, f) and the winter (DJF) temperature time series for 160 stations in China (shaded: significance and contour: correlation coefficient values) and the composite differences in the temperatures (dots; unit: C) between selected low BKI years and high BKI years. +e upper panels are for the OBKI, and the bottom panels are for the DBKI. Both the shaded areas and the dots indicate values that are statistically significant with significance exceeding 0.1 based on the results of t-tests. +e values in the color bar represent the significance levels of the t-test. positive correlation with the SAT than the detrended ASI. EOF2 is characterized by an opposite variation be- Comparatively, the correlation between the ASI and the SAT tween Northeast China and the other regions in China is notably lower than the correlation between the BKI and (Figure 4(b)). +e persistent positive value of PC2 sug- PC1. +is implies that the BK is the key region in the Arctic gests a cold anomaly in the western and southern parts of in which the change in the sea ice cover may play a critical China before the late 1990s, resulting in warmer condi- role in the winter SAT anomalies in China. tions in these areas. However, neutral positive 0.4 0.26 0.4 0.4 0.31 0.4 0.31 0.26 0.4 0.26 Advances in Meteorology 7 50°N 50°N 1.2 0.2 40°N 40°N 0.8 0.6 30°N 30°N 20°N 20°N 70°E 80°E 90°E 100°E 110°E 120°E 130°E 140°E 70°E 80°E 90°E 100°E 110°E 120°E 130°E 140°E –1 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1 –1 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1 (a) (b) 2.5 2 2 1.5 1.5 1 1 0.5 0.5 0 0 –0.5 –0.5 –1 –1 –1.5 –1.5 –2 –2 –2.5 –2.5 –3 1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015 (c) (d) Figure 4: Spatial patterns of the first (a) and the second (b) EOF modes of the winter surface air temperature anomalies at 160 stations in China from 1979 to 2018 (unit: C); the corresponding normalized time coefficients for the first (c) and the second (d) EOF modes. Table 1: Correlation coefficients between the month-to-month detrended (original) BKI, ASI, PC1, and PC2. BKI ASI EOF 9 10 11 9 10 11 ∗ ∗ ∗ PC1 0.48 (0.27) 0.42 (0.18) 0.19 (0.00) 0.40 (0.22) 0.39 (0.29) 0.18 (0.23) PC2 0.03 (− 0.16) 0.18 (0.29) 0.39 (0.47 ) 0.16 (0.33) 0.08 (0.27) 0.19 (0.31) +e bold font and asterisks indicate the significance exceeding 0.05 and 0.01, respectively. correlations are indicated between PC2 and the sea ice 5. Sea Ice Loss-Associated Circulation cover in both the BK and the Arctic except for the BKI in November. To explain how the preceding sea ice variation induces +e above results show a decreasing relationship be- winter SAT anomalies in China, composite differences be- tween the month-to-month variability of the BKI and the tween the sea-level pressure (SLP) and the 850 hPa flow winter SAT in China, which suggests that more attention based on low-high OBKI (DBKI) values in September are should be paid to the preceding September sea ice cover in presented in Figures 5(a) and 5(c). +e difference field shows the BK instead of the signals near October and November, a positive anomalous winter pressure system in the north- which is when we focus on the seasonal prediction of western part of the Eurasian continent that is suggestive of winter SAT anomalies in China. Moreover, the detrended the northwestward expansion and intensification of the Cold BKI time series presents a higher correlation than the Siberian High following the preceding loss of sea ice in the original time series, implying that the interannual vari- BK. Correspondingly, a massive anticyclone in the lower ability in the sea ice cover may play a more critical role in troposphere is present that is superimposed on the high- its association with the SAT in China compared with its pressure system. +is anomalous circulation pattern facili- linear trend component. tates the invasion of cold air from high latitudes to East Asia 8 Advances in Meteorology 80°N 80°N 200 400 70°N 70°N 1 60°N 60°N –1 –2 50°N 50°N –1 –2 –1 40°N 100 40°N 30°N 30°N –100 20°N 20°N 0 15°E 30°E 45°E 60°E 75°E 90°E 105°E 120°E 135°E 150°E 0 15°E 30°E 45°E 60°E 75°E 90°E 105°E 120°E 135°E 150°E (a) (b) 80°N 80°N 2 5 70°N 70°N 60°N 60°N 300 –1 –3 50°N 50°N –1 –2 40°N 40°N –1 –2 30°N 30°N –1 20°N 20°N 0 15°E 30°E 45°E 60°E 75°E 90°E 105°E 120°E 135°E 150°E 0 15°E 30°E 45°E 60°E 75°E 90°E 105°E 120°E 135°E 150°E (c) (d) − 1 Figure 5: Composite differences for the (a) SLP (shaded and contour; unit: Pa), wind (vector; unit: m·s ) at 850 hPa, and (b) surface air temperature (shaded and contour; unit: C) between selected low OBKI years and high OBKI years in September. (c, d) +e same as (a) and (b) but for the composite differences between the selected low DBKI years and high DBKI years in September. +e shading denotes significance exceeding 0.1. +e black vectors indicate winds with a northerly component over the area east of 100 E. and the weakening of warm midlatitude westerlies from the Table 2: Correlation coefficients between the DBKI in September North Atlantic that are instrumental to the cold advection. and the circulation indices in the winter. +is corresponds to large areas of cold anomalies in the midlatitudes of Eurasia as well as East Asia (Figures 5(b) and Circulation indices Correlation coefficient 5(d)). Comparatively, the significant areas of the SAT be- Westerly index 0.42 tween the low-high DBKI in September in Figure 5(d) ex- Eurasian blocking high frequency − 0.65 tend more southeastward to the eastern part of China, the Eurasian anticyclone intensity − 0.55 Korean Peninsula, Japan, and the Northwest Pacific than the Siberian High intensity − 0.45 DBKI in Figure 5(b). A substantial positive high-pressure East Asian winter monsoon index − 0.44 AO index 0.21 system accompanied by a low-pressure system over the Northwest Pacific intensifies the land-sea pressure gradient. +e bold font indicates significance exceeding 0.01. A northerly component correspondingly prevails in the area east of 100 E on the Eurasian continent, which contributes to an intensification of the East Asian winter monsoon. correlate positively with each other at 0.42, indicating fa- However, the northerly component of the low-high OBKI is vorable weakening westerly conditions and a meridional mainly present over the northeastern part of Eurasia. +e circulation pattern following the loss of sea ice in the BK. +e associated response of the intensified East Asian winter significant anticorrelation between the blocking frequency, anticyclone intensity, and Siberian High index and the DBKI monsoon to the low-high DBKI contributes to colder conditions in China compared with the response to the suggest that the retreat of the sea ice cover facilitates the maintenance and intensification of synoptic fluctuations, OBKI in September. +erefore, the following analysis is mainly based on the DBKI in September. such as blocking highs and surface cold highs, which lead to To explore the wintertime atmospheric response over an accumulated strengthening of the semipermanent Sibe- Eurasia to the preceding variation of sea ice, we calculated rian High. Correspondingly, an increase in the land-sea the correlation coefficients between the DBKI in September pressure gradient over East Asia suggests an intensified East and the circulation indices in winter. As shown in Table 2, Asian monsoon that leads to large-scale cold anomalies in most of the circulation indices are closely related to the China. On the contrary, although the AO is well represented September DBKI with significance exceeding 0.01 except for with the zonal mean status of midlatitude westerlies [38], the the AO index. In particular, the DBKI and the westerly index relationship of the westerly index with the DBKI is notably Advances in Meteorology 9 closer to the relationship between the AO and the DBKI. air temperature in the BK shows clearly persistent warming +is implies a local connection of the preceding sea ice in the from September to February and potentially results in a BK with the midlatitude climate anomalies. Kug et al. [14] decrease of the meridional temperature gradient in the also documented different local responses of the variations winter. Coincidently, the autumn heat content of the upper of the sea ice cover in the BK and in the East Siberian and ocean in the BK presents a significantly warm difference Chukchi seas; the variation of the sea ice cover in the BK is between low and high DBKI (Figures 7(a)–7(c)). Moreover, associated well with the winter SAT anomalies over the a slightly warmer condition is induced by low-high Sep- Eurasian continent, while the variation of the sea ice cover in tember DBKI (S-DBKI) compared to the October DBKI (O- the East Siberian and Chukchi seas is mainly linked with the DBKI), and the contrast of oceanic warming is even more SAT anomalies over North America. clear between the S-DBKI and the November DBKI (N- +e reduction in sea ice has a direct connection to the DBKI). +e seasonal sea ice melting from May to September warming of the air column and modifies the meridional opens a large portion of the Arctic Ocean, allowing it to thickness gradient over mid- to high-latitude regions, absorb sunlight during the warm season [40]. +erefore, the leading to the change in midlatitude weather [8, 12]. To retreat of the autumn sea ice cover in the BK associates with examine the linkage between sea ice loss and midlatitude more heat storage in the upper BK, especially in September. circulation, the regional means of the air temperature Correspondingly, the subsequent December-January mean ° ° ° ° profiles over the BK (30 E–100 E, 75 N–85 N, referred to as surface sensible + latent heat flux and upward longwave the north region (NR)) and the midlatitudes of Eurasia radiation show a clear increase in the BK (Figures 7(d)–7(f)). ° ° ° ° (30 E–100 E, 40 N–50 N, referred to as the south region +e positive anomalous ocean-to-atmosphere energy in- (SR)) were calculated, and their composite differences dicates an additional release of oceanic heat storage in the between low-high sea ice years are shown in Figure 6. A BK to the atmosphere associated with the preceding low continuous warming in the NR from September (0 yr) to autumn sea ice cover. +e winter heat anomalies in the BK February (+1 yr) is indicated in the lower troposphere, but and their surroundings are most pronounced to be asso- it decreases with an increase in the height (Figure 6(a)). ciated with the S-DBKI compared to the O-DBKI and Lower-troposphere warming is typically induced by the N-DBKI. release of heat flux from the open ocean to the atmosphere Figure 8 shows the composite differences between the due to the reduction in sea ice [39]. A clear difference in the December-January mean 500 hPa geopotential height warming in the NR is observed between the late autumn (Z500) and the 850 hPa flow (V850) based on low-high and winter. During October and November, the anomalous DBKI values from September to November. A pronounced warming is mainly displayed below 850 hPa with a low high Z500 center is shown in the BK and extends southward significance, and a weak cooling can also be found in the to the Ural Mountains associated with low S-DBKI upper air in November. However, substantial deep (Figure 8(a)). Similar Z500 anomaly patterns also present warming occurs in the midtroposphere and low tropo- through the whole troposphere with a barotropic vertical sphere in the winter with a maximum value of 3.2 C near structure (figure not shown) and link with the amplification the surface. Moreover, dominant cooling is evident in the tropospheric warming induced by sea ice loss. Furthermore, the anomalous southerlies from the high latitudes of SR, especially in January and February (Figure 6(b)). In addition, the September DBKI is associated more closely Northern Atlantic drive warm advection into the BK and facilitate a further amplification warming. +e warming over with midlatitude SAT variations in the winter than in the autumn. Furthermore, the composite differences of di- the BK links with the release of oceanic heat storage to the abetic heating in the NR (SR) from December (0 yr) to atmosphere and associated anomalous warm advection over February (+1 yr) between low-high sea ice years are 39.46 the Barents Sea and cold advection over the midlatitude of − 2 − 2 − 2 (9.04) w·m , 34.14 (− 9.55) w·m , and 29.74(− 7.68) w·m , the Eurasia. A notable difference in temperature between the respectively. +is suggests that the retreat of the sea ice in BK and the midlatitude of Eurasia results in the significant the BK in autumn favors the mid- and lower-tropospheric weakened meridional temperature gradient. +is decelerates warming and thus weakens the meridional temperature the westerlies and leads to more active surface anticyclone between the high latitude and the midlatitude. +e thermal and blocking (Table 2) and anticyclonic vorticity forcing to their north [41]. Correspondingly, large-scale anticyclonic conditions in the BK may play a driving role in the changes in the meridional temperature gradient and the associated circulation is shown over the northwest of Eurasia. As discussed in Section 5, the anomalous anticyclonic circu- thickness gradient. lation over the Ural Mountains and BK leads to more Ural blocking, a deeper East Asian trough, and more intense cold 6. Month-to-Month Oceanic Heat Storage air invasion into the East Asian region. In contrast, the low O-DBKI and N-DBKI associated Z500 anomalies display a Contrast and Associated northwestward-displaced positive center over the west of Subsequent Warming Greenland (Figures 8(b) and 8(c)). Cold advection from the In this section, we will address the reason why the DBKI in east Siberian sector of Arctic is weaker compared to S-DBKI- September could be a better indicator of the SAT anomalies associated Z500 anomalies (Figure 8(a)). Consequently, the S-DBKI shows a higher correlation with winter SAT in China compared to the DBKI in the following October and November. As discussed above, the lower-troposphere anomalies in China. 10 Advances in Meteorology hPa hPa hPa 500 500 500 600 600 600 700 700 700 775 775 775 850 850 850 0.8 925 0.8 925 925 1.6 –0.8 1000 1000 1000 Sep Oct Nov Dec Jan Feb Sep Oct Nov Dec Jan Feb Sep Oct Nov Dec Jan Feb (a) (b) (c) ° ° ° ° ° Figure 6: Composite differences for (a) vertical profiles of the temperature (shaded; unit: C) in the BK region/NR (30 E–100 E, 75 N–85 N) from September to the following February between selected low DBKI years and high DBKI years in September. (b, c) +e same as (a) but for ° ° ° ° the midlatitude region/SR (30 E–100 E, 40 N–50 N) and the differences between the SR and the NR, respectively. +e green dot denotes the values that are statistically significant with significance exceeding 0.1 based on the results of t-tests. 90°W 90°W 90°W 60°W 120°W 60°W 120°W 60°W 120°W 30°W 150°W 150°W 30°W 150°W 30°W 0° 180° 0° 180° 0° 180° 150°E 150°E 150°E 30°E 30°E 30°E 120°E 120°E 60°E 60°E 60°E 120°E 90°E 90°E 90°E Contour from –0.8 to 1.4 by 0.244 Contour from –0.8 to 1.4 by 0.244 Contour from –0.8 to 1.4 by 0.244 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1.1 1.4 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1.1 1.4 –0.8 –0.6 –0.4 –0.2 0.2 0.4 0.6 0.8 1.1 1.4 (a) (b) (c) 90°W 90°W 90°W 60°W 120°W 60°W 120°W 60°W 120°W 150°W 150°W 150°W 30°W 30°W 30°W 0° 180° 0° 180° 0° 180° 30°E 150°E 30E 150°E 30°E 150°E 60°E 120°E 60°E 120°E 60°E 120°E 90°E 90°E 90°E Contour from –100 to 100 by 20 Contour from –100 to 80 by 20 Contour from –140 to 140 by 20 –20 –10 10 20 30 40 50 60 –20 –10 10 20 30 40 50 60 –20 –10 10 20 30 40 50 60 (d) (e) (f ) Figure 7: Composite differences for the SON mean upper ocean heat content (shaded) between selected low and high DBKI in September (a), October (b), and November (c) (unit: K). (d–f) +e same as (a), (b), and (c) but for DJF mean upward longwave radial (shaded) and − 2 sensible + latent heat flux (contour) (unit: w·m ). +e dotted areas denote the values that are statistically significant with significance exceeding 0.1 based on the results of t-tests for heat content in (a), (b), and (c) and longwave radiation in (d), (e), and (f). –4.8 –4 –3.2 –2.4 –1.6 –0.8 0.8 1.6 2.4 3.2 4.8 –4.8 –4 –3.2 –2.4 –1.6 –0.8 0.8 1.6 2.4 3.2 4.8 –4.8 –4 –3.2 –2.4 –1.6 –0.8 0.8 1.6 2.4 3.2 4.8 Advances in Meteorology 11 90°W 90°W 90°W 60°W 120°W 60°W 120°W 60°W 120°W 150°W 150°W 30°W 30°W 30°W 150°W 0° 0° 0° 180° 180° 180° 150°E 150°E 30°E 30°E 150°E 30°E 6 6 120°E 60°E 120°E 60°E 60°E 120°E 90°E 90°E 90°E –80 –60 –40 –20 –10 10 20 40 60 80 –80 –60 –40 –20 –10 10 20 40 60 80 –80 –60 –40 –20 –10 10 20 40 60 80 (a) (b) (c) − 1 Figure 8: +e same as (d), (e), and (f) in Figure 7 but for DJF mean Z500 (shaded) (unit: gpm) and V850 (vector) (unit: m·s ). +e dotted areas denote the values that are statistically significant with significance exceeding 0.1 based on the results of t-tests for Z500. +is regional tropospheric warming results in a 7. Conclusions higher barotropic positive height anomaly over the +e variation of the autumn Arctic sea ice area and the re- Ural Mountains, and thus, more active cold advec- lationship between the month-to-month sea ice and the winter tion from the high latitude affects East Asia. temperature anomalies in China are investigated in this study. +e conclusions based on the main results are as follows: Data Availability (1) +e sea ice cover in the BK during the autumn season +e data used to support the findings of this study are in- has shown a substantial decrease during the past cluded within the article: (1) the European Centre for three decades. +e region of accelerating sea ice Medium-Range Weather Forecasts (ECMWF) reanalysis reduction since 2005 agrees well with the Arctic dataset (ERA-Interim) from 1979 to 2018 is used for deriving superimposed with a large interannual variability. data on the daily and monthly air temperatures, sea-level (2) +e retreat of autumn sea ice in the BK is significantly pressure, geopotential height, and horizontal wind; (2) associated with large-scale negative temperature monthly mean sea ice concentrations are derived from the anomalies in the following winter in China, and this Met Office Hadley Centre’s sea ice and sea surface tem- retreat correlates well with the dominant sign-con- perature (HadiSST 1.1) dataset; (3) the monthly mean heat sistent SAT mode (EOF1) in China, except for the content is derived from the vertical average of potential Tibetan Plateau. However, the ice-SAT lag correlation temperature (109.8 m above; 11 levels) from the ECMWF shows a notable month-to-month diversity in which ocean reanalysis system ORAS4; and (4) observed monthly the linkage between the sea ice in the BK in September mean surface air temperatures (SATs) are from the National and the winter temperature is stronger than that in Climate Center of China, including 160 stations. both October and November. Moreover, the detrended sea ice cover in the BK as a potential Conflicts of Interest predictor is associated more closely with the SAT in China than the original sea ice cover in the BK. +e authors declare that there are no conflicts of interest (3) An anomalous positive surface pressure is exhibited regarding the publication of this paper. over the northwestern part of Eurasia in the winter that is linked to the decreasing sea ice in the BK region Acknowledgments in the preceding September. +is surface pressure favors the persistence and intensification of synoptic +is work was supported jointly by the National Basic perturbations, such as blocking highs and surface cold Research Program of China (Grant no. 2015CB953904) and highs, as well as the intensification of the Siberian the National Natural Science Foundation of China (Grant High and East Asian winter monsoon. +ese favorable nos. 41975073, 41575081, and 41741005). conditions ultimately contribute to the formation of large-scale winter cold anomalies in China. References (4) Compared to low O-DBKI and N-DBKI, a more oceanic heat storage in the upper BK is induced by [1] T. R. Karl, A. Arguez, B. Huang et al., “Possible artifacts of low S-DBKI leading to a larger heat release to tro- data biases in the recent global surface warming hiatus,” pospheric atmosphere by surface heat flux in the BK. Science, vol. 348, no. 6242, pp. 1469–1472, 2015. 12 Advances in Meteorology [2] X. Zhang, C. Lu, and Z. Guan, “Weakened cyclones, in- Meteorologica Sinica, vol. 16, no. 4, pp. 34–40, 1994, in tensified anticyclones and recent extreme cold winter weather Chinese. events in Eurasia,” Environmental Research Letters, vol. 7, [20] X. M. Shi, Y. W. Sun, and J. L. Sun, “+e impact of zonal no. 4, Article ID 044044, 2012. temperature gradient in the Arctic Ocean on the summer [3] C. H. Lu, S. Q. Xie, Y. J. Qin et al., “Recent intensified winter precipitation over Northeast China,” Periodical of Ocean coldness in the mid-high latitudes of Eurasia and its re- University of China, vol. 44, no. 2, pp. 11–16, 2014, in Chinese. lationship with daily extreme low temperature variability,” [21] H. J. Wang, H. P. Chen, and J. Liu, “Arctic sea ice decline Advances in Meteorology, vol. 2016, Article ID 3679291, intensified haze pollution in Eastern China,” Atmospheric and 11 pages, 2016. Oceanic Science Letters, vol. 8, no. 1, pp. 1–9, 2015. [4] L. Yang and B. Wu, “Interdecadal variations of the East Asian [22] B. Y. Wu, R. H. Huang, and D. Y. Gao, “+e impact of variation winter surface air temperature and possible causes,” Chinese of sea-ice extent in the Kara Sea and the Barents Seas in winter on Science Bulletin, vol. 58, no. 32, pp. 3969–3977, 2013. the winter monsoon over East Asia,” Chinese Journal of At- [5] L. Wang and W. Chen, “+e East Asian winter monsoon: Re- mospheric Sciences, vol. 23, no. 3, pp. 267–275, 1999, in Chinese. amplification in the mid-2000s,” Chinese Science Bulletin, [23] X. H. Xie and X. Q. Yang, “Interannual relationship between vol. 59, no. 4, pp. 430–436, 2014. Arctic sea ice concentration anomaly and China air tem- [6] J. A. Francis, “Why are Arctic linkages to extreme weather still perature variation during winter,” Journal of Nanjing Uni- up in the air?,” Bulletin of the American Meteorological Society, versity (Natual Sciences), vol. 42, no. 6, pp. 549–561, 2006. vol. 98, no. 12, pp. 2551–2557, 2017. [24] B. Y. Wu, J. Z. Su, and R. H. Zhang, “Effects of autumn-winter [7] J. Overland, “Potential arctic change through climate am- Arctic sea ice on winter Siberian High,” Chinese Science plification processes,” Oceanography, vol. 24, no. 3, pp. 176– Bulletin, vol. 56, no. 30, pp. 3220–3228, 2011. 185, 2011. [25] F. Li, H. J. Wang, and Y. Q. Gao, “Change in sea ice cover is [8] J. A. Francis and S. J. Vavrus, “Evidence linking Arctic am- responsible for non-uniform variation in winter temperature plification to extreme weather in mid-latitudes,” Geophysical over East Asia,” Atmospheric and Oceanic Science Letters, Research Letters, vol. 39, no. 6, Article ID L06801, 2012. vol. 8, no. 6, pp. 376–382, 2015. [9] B.-M. Kim, S.-W. Son, S.-K. Min et al., “Weakening of the [26] C. Y. Zhu, F. Huang, Y. H. Shi et al., “Spatial-temporal stratospheric polar vortex by Arctic sea-ice loss,” Nature patterns of the cold surge events in China in recent 50 years Communications, vol. 5, no. 1, 2014. and its relationship with Arctic sea ice,” Periodical of Ocean [10] E. A. Barnes, “Revisiting the evidence linking Arctic ampli- University of China, vol. 44, no. 12, pp. 12–20, 2014, in fication to extreme weather in midlatitudes,” Geophysical Chinese. Research Letters, vol. 40, no. 17, pp. 4734–4739, 2013. [27] C. Li and Q. Zhang, “January temperature anomalies over [11] C. Deser and H. Teng, “Evolution of Arctic sea ice concen- Northeast China and precursors,” Chinese Science Bulletin, tration trends and the role of atmospheric circulation forcing vol. 58, no. 6, pp. 671–677, 2013. 1979–2007,” Geophysical Research Letters, vol. 35, no. 2, [28] D. W. Wang and X. Q. Yang, “Temporal and spatial patterns Article ID L02504, 2008. of Arctic sea ice variations,” Acta Meteorologica Sinica, vol. 60, [12] J. E. Overland and M. Wang, “Large-scale atmospheric cir- no. 2, pp. 129–138, 2002. culation changes are associated with the recent loss of Arctic [29] J. H. He, F. M. Wu, and L. Qi, “Decadal/interannual linking sea ice,” Tellus A: Dynamic Meteorology and Oceanography, between autumn Arctic sea ice and following winter Eurasian vol. 62, no. 1, pp. 1–9, 2010. air temperature,” Chinese Journal of Geophysics, vol. 58, no. 4, [13] M. Mori, M. Watanabe, H. Shiogama, J. Inoue, and pp. 1089–1102, 2015, in Chinese. M. Kimoto, “Robust Arctic sea-ice influence on the frequent [30] D. P. Dee, S. M. Uppala, A. J. Simmons et al., “+e ERA- Eurasian cold winters in past decades,” Nature Geoscience, Interim reanalysis: configuration and performance of the data vol. 7, no. 12, pp. 869–873, 2014. assimilation system,” Quarterly Journal of the Royal Meteo- [14] J. S. Kug, J. H. Jeong, Y. S. Jang et al., “Two distinct influences rological Society, vol. 137, no. 656, pp. 553–597, 2011. of Arctic warming on cold winters over North America and [31] N. A. Rayner, D. E. Parker, and E. B. Horton, “Global analyses East Asia,” Nature Geoscience, vol. 8, no. 10, Article ID of sea surface temperature, sea ice, and night marine air NGEO2517, pp. 759-762, 2015. temperature since the late nineteenth century,” Journal of [15] P. Ruggieri, R. Buizza, and G. Visconti, “On the link between Geophysical Research, vol. 108, no. D14, 2003. Barents-Kara sea ice variability and European blocking,” [32] M. A. Balmaseda, K. Mogensen, and A. T. Weaver, “Evalu- Journal of Geophysical Research: Atmospheres, vol. 121, no. 10, ation of the ECMWF ocean reanalysis system ORAS4,” pp. 5665–5679, 2015. Quarterly Journal of the Royal Meteorological Society, vol. 139, [16] Z. Lu, ¨ S. He, F. Li, and H. Wang, “Impacts of the autumn no. 674, pp. 1132–1161, 2013. Arctic sea ice on the intraseasonal reversal of the winter [33] D. Y. Gong and C. H. Ho, “+e Siberian High and climate Siberian High,” Advances in Atmospheric Sciences, vol. 36, change over middle to high latitude Asia,” 9eoretical and no. 2, pp. 173–188, 2019. Applied Climatology, vol. 72, no. 1-2, pp. 1–9, 2002. [17] Q. Xie and S. S. Huang, “Study of influence of equatorial SST [34] J. G. Charney, J. Shukla, and K. C. Mo, “Comparison of a over Mid-East Pacific ocean and Arctic sea-ice anomalies to barotropic blocking theory with observation,” Journal of the general circulation,” Scientia Meteorologica Sinica, vol. 10, Atmospheric Sciences, vol. 38, no. 4, pp. 762–779, 1981. no. 4, pp. 325–338, 1990. [35] N. Shi and Y. S. Yang, “Main characteristics of east Asian [18] B. Wu, D. Gao, and R. Huang, “ENSO events and interannual summer/winter intermonsoon index for 1873∼1996,” Journal variations of winter sea-ice in the Greenland, the Kara and the Barents Seas,” Chinese Science Bulletin, vol. 42, no. 16, of Nanjing Institute of Meteorology, vol. 21, no. 2, pp. 208–214, pp. 1382–1385, 1997. 1998, in Chinese. [19] X. Q. Yang, Q. Xie, and S. S. Huang, “Modelling influence of [36] J. Inoue and E. H. Masatake, “+e role of Barents sea ice in the Arctic sea-ice anomalies to Asian summer monsoon,” Acta wintertime cyclone track and emergence of a warm-Arctic Advances in Meteorology 13 cold-Siberian anomaly,” Journal of Climate, vol. 25, no. 7, [37] G. R. North, T. L. Bell, R. F. Cahalan, and F. J. Moeng, “Sampling errors in the estimation of empirical orthogonal functions,” Monthly Weather Review, vol. 110, no. 7, pp. 699–706, 1982. [38] D. W. J. +ompson and J. M. Wallace, “Annular modes in the extratropical circulation. Part I: month-to-month variability,” Journal of Climate, vol. 13, no. 5, pp. 1000–1016, 2000. [39] J. A. Screen and I. Simmonds, “+e central role of diminishing sea ice in recent Arctic temperature amplification,” Nature, vol. 464, no. 7293, pp. 1334–1337, 2010. [40] A. Dai, D. Luo, M. Song et al., “Amplification is caused by sea- ice loss under increasing CO ,” Nature Communications, vol. 10, no. 121, pp. 1–13, 2019. [41] N.-C. Lau, “Variability of the observed midlatitude storm tracks in relation to low-frequency changes in the circulation pattern,” Journal of the Atmospheric Sciences, vol. 45, no. 19, pp. 2718–2743, 1988. International Journal of The Scientific Advances in Advances in Geophysics Chemistry Scientica World Journal Public Health Hindawi Hindawi Hindawi Hindawi Publishing Corporation Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 Journal of Environmental and Public Health Advances in Meteorology Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Applied & Environmental Journal of Soil Science Geological Research Hindawi Volume 2018 Hindawi www.hindawi.com www.hindawi.com Volume 2018 International Journal of International Journal of Agronomy Ecology International Journal of Advances in International Journal of Forestry Research Microbiology Agriculture Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 International Journal of Journal of Journal of International Journal of Biodiversity Archaea Analytical Chemistry Chemistry Marine Biology Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Advances in Meteorology – Hindawi Publishing Corporation

Published: Dec 12, 2019

There are no references for this article.