• Alexeev, V. A., V. V. Ivanov, R. Kwok, and L. H. Smedsrud, 2013: North Atlantic warming and declining volume of arctic sea ice. Cryosphere Discuss., 7, 245265, doi:10.5194/tcd-7-245-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ärthun, M., and T. Eldevik, 2016: On anomalous ocean heat transport toward the Arctic and associated climate predictability. J. Climate, 29, 689704, doi:10.1175/JCLI-D-15-0448.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Burt, M. A., D. A. Randall, and M. D. Branson, 2016: Dark warming. J. Climate, 29, 705719, doi:10.1175/JCLI-D-15-0147.1.

  • Cavalieri, D. J., and C. L. Parkinson, 2012: Arctic sea ice variability and trends, 1979–2010. Cryosphere, 6, 881889, doi:10.5194/tc-6-881-2012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chylek, P., C. K. Folland, G. Lesins, M. K. Dubey, and M. Wang, 2009: Arctic air temperature change amplification and the Atlantic Multidecadal Oscillation. Geophys. Res. Lett., 36, L14801, doi:10.1029/2009GL038777.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cohen, J., and Coauthors, 2014: Recent Arctic amplification and extreme mid-latitude weather. Nat. Geosci., 7, 627637, doi:10.1038/ngeo2234.

  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Diao, Y., J. Li, and D. Luo, 2006: A new blocking index and its application: Blocking action in the Northern Hemisphere. J. Climate, 19, 48194839, doi:10.1175/JCLI3886.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fang, Z., and J. M. Wallace, 1994: Arctic sea ice variability on a timescale of weeks and its relation to atmospheric forcing. J. Climate, 7, 18971914, doi:10.1175/1520-0442(1994)007<1897:ASIVOA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francis, J. A., and E. Hunter, 2006: New insight into the disappearing Arctic sea ice. Eos, Trans. Amer. Geophys. Union, 87, 509524, doi:10.1029/2006EO460001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francis, J. A., and E. Hunter, 2007: Drivers of declining sea ice in the Arctic winter: A tale of two seas. Geophys. Res. Lett., 34, L17503, doi:10.1029/2007GL030995.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francis, J. A., E. Hunter, J. R. Key, and X. Wang, 2005: Clues to variability in Arctic minimum sea ice extent. Geophys. Res. Lett., 32, L21501, doi:10.1029/2005GL024376.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ghatak, D., and J. Miller, 2013: Implications for Arctic amplification of changes in the strength of the water vapor feedback. J. Geophys. Res. Atmos., 118, 75697578, doi:10.1002/jgrd.50578.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lim, Y. K., 2015: The east Atlantic/west Russia (EA/WR) teleconnection in the North Atlantic: Climate impact and relation to Rossby wave propagation. Climate Dyn., 44, 32113222, doi:10.1007/s00382-014-2381-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lindsay, R. W., and J. Zhang, 2005: The thinning of Arctic sea ice, 1988–2003: Have we passed a tipping point? J. Climate, 18, 48794894, doi:10.1175/JCLI3587.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Luo, D., A. Lupo, and H. Wan, 2007: Dynamics of eddy-driven low-frequency dipole modes. Part I: A simple model of North Atlantic Oscillations. J. Atmos. Sci., 64, 338, doi:10.1175/JAS3818.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Luo, D., Y. Xiao, Y. Yao, A. Dai, I. Simmonds, and C. Franzke, 2016a: Impact of Ural blocking on winter warm Arctic–cold Eurasian anomalies. Part I: Blocking-induced amplification. J. Climate, 29, 39253947, doi:10.1175/JCLI-D-15-0611.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Luo, D., Y. Xiao, Y. Diao, A. Dai, C. Franzke, and I. Simmonds, 2016b: Impact of Ural blocking on winter warm Arctic–cold Eurasian anomalies. Part II: The link to the North Atlantic Oscillation. J. Climate, 29, 39493971, doi:10.1175/JCLI-D-15-0612.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Miles, M. W., D. V. Divine, T. Furevik, E. Jansen, M. Moros, and A. E. J. Ogilvie, 2014: A signal of persistent Atlantic multidecadal variability in Arctic sea ice. Geophys. Res. Lett., 41, 463469, doi:10.1002/2013GL058084.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Overland, J., J. Francis, R. Hall, E. Hanna, S. Kim, and T. Vihma, 2015: The melting Arctic and midlatitude weather patterns: Are they connected? J. Climate, 28, 79177932, doi:10.1175/JCLI-D-14-00822.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Park, D.-S., S. Lee, and S. B. Feldstein, 2015: Attribution of the recent winter sea-ice decline over the Atlantic sector of the Arctic Ocean. J. Climate, 28, 40274033, doi:10.1175/JCLI-D-15-0042.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Park, H. S., S. Lee, S. W. Son, S. B. Feldstein, and Y. Kosaka, 2015: The impact of poleward moisture and sensible heat flux on Arctic winter sea ice variability. J. Climate, 28, 50305039, doi:10.1175/JCLI-D-15-0074.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Peings, Y., and G. Magnusdottir, 2014: Forcing of the wintertime atmospheric circulation by the multidecadal fluctuations of the North Atlantic Ocean. Environ. Res. Lett., 9, 034018, doi:10.1088/1748-9326/9/3/034018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rigor, I. G., J. M. Wallace, and R. L. Colony, 2002: Response of sea ice to the Arctic Oscillation. J. Climate, 15, 26482663, doi:10.1175/1520-0442(2002)015<2648:ROSITT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Screen, J. A., and I. Simmonds, 2010: The central role of diminishing sea ice in recent Arctic temperature amplification. Nature, 464, 13341337, doi:10.1038/nature09051.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Screen, J. A., and Coauthors, 2010: Increasing fall–winter energy loss from the Arctic Ocean and its role in Arctic temperature amplification. Geophys. Res. Lett., 37, L16707, doi:10.1029/2010GL044136.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Simmonds, I., 2015: Comparing and contrasting the behaviour of Arctic and Antarctic sea ice over the 35 year period 1979–2013. Ann. Glaciol., 56, 1828, doi:10.3189/2015AoG69A909.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sorokina, S. A., C. Li, J. J. Wettstein, and N. G. Kvamstø, 2016: Observed atmospheric coupling between Barents Sea ice and the warm-Arctic cold-Siberian anomaly pattern. J. Climate, 29, 495511, doi:10.1175/JCLI-D-15-0046.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sorteberg, A., and B. Kvingedal, 2006: Atmospheric forcing on the Barents Sea winter ice extent. J. Climate, 19, 47724784, doi:10.1175/JCLI3885.1.

  • Tibaldi, S., and F. Molteni, 1990: On the operational predictability of blocking. Tellus, 42A, 343365, doi:10.1034/j.1600-0870.1990.t01-2-00003.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Walsh, J. E., 2014: Intensified warming of the Arctic: Causes and impacts on middle latitudes. Global Planet. Change, 117, 5263, doi:10.1016/j.gloplacha.2014.03.003.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, M., and J. E. Overland, 2009: A sea ice free summer Arctic within 30 years? Geophys. Res. Lett., 36, L07502, doi:10.1029/2009GL037820.

  • Woods, C., and R. Caballero, 2016: The role of moist intrusions in winter Arctic warming and sea ice decline. J. Climate, 29, 44734485, doi:10.1175/JCLI-D-15-0773.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Woods, C., R. Caballero, and G. Svensson, 2013: Large-scale circulation associated with moisture intrusions into the Arctic during winter. Geophys. Res. Lett., 40, 47174721, doi:10.1002/grl.50912.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • View in gallery

    Time series of DJF-mean SIC anomalies (red dashed line denotes a detrending) over (a) the whole Arctic and (b) BKS (65°–80°N, 0°–90°E) during 1979–2011; and linear regression of the winter-mean 500-hPa geopotential height (contour interval: 50 gpm) and SAT (color shading) anomalies against the detrended DJF-mean SIC index (multiplied by −1) over (c) the whole Arctic and (d) BKS during 1979–2011. The dotted areas show values that are statistically significant at the p < 0.05 level for the Student’s t test.

  • View in gallery

    Linear regression of the DJF-mean 500-hPa geopotential height (contour interval: 50 gpm) and SAT (color shading) anomalies against the detrended DJF-mean SIC index (multiplied by −1) over (a) the whole Arctic and (b) BKS during 1987–2011. The dotted areas show values that are statistically significant at the p < 0.05 level for the Student’s t-test.

  • View in gallery

    Time series of the normalized winter-mean (a) SST over the BKS (the red dashed line denotes a detrending) and (b) UB frequency (days) from 1979–2011 (red dashed line denotes a 5-yr running mean); and linear regression of the DJF-mean (c) SIC against the SST and (d) SIC against the UB frequency during 1979–2011. The color shading denotes the region above the 90% confidence level for the Student’s t test.

  • View in gallery

    Lagged regressions of the daily (a) 500-hPa geopotential height, (b) SAT, and (c) SIC anomalies against the daily TM strength index for the 1979–2011 winters (DJF). The color shading denotes the regions above the 95% confidence level for the two-sided Student’s t test. Lag 0 denotes the day of the UB peak.

  • View in gallery

    Lagged regressions of the daily (a) vertically integrated moisture flux convergence (multiplied by L), (b) TCW, and (c) downward IR anomalies against the daily TM strength index for the 1979–2011 winters. The color shading in (b) and (c) denotes the region above the 95% confidence level for the two-sided Student’s t test, whereas the color shading in (a) represents the region above the 90% confidence level. Lag 0 denotes the day of the UB peak.

  • View in gallery

    Pattern correlations between the daily 500-hPa geopotential height (Hgt_500mb), moisture flux convergence (Mois_con), TCW (liquid water plus ice), and downward IR and SAT anomalies and the same variables obtained by lagged regression against the daily TM strength index for the 1979–2011 winters. The sea ice pattern correlation (red line) is derived by performing the lagged regressions of the daily SIC and the same variable obtained by regressing against the wintertime UB frequency (see Fig. 3b). Lag 0 denotes the day of the UB peak. The dots show values that are statistically significant at the p < 0.05 level for the Student’s t test.

  • View in gallery

    (a) Time series of the SAT anomaly averaged over the BKS (red dashed curve denotes a detrending) during 1979–2011 and regressed 300-hPa zonal wind fields against the detrended (b) SAT in (a) and (c) SST anomalies averaged over the BKS in Fig. 3a. The dotted areas show values that are statistically significant at the p < 0.05 level for the two-sided Student’s t test.

  • View in gallery

    Schematic diagram of the EA/WR pattern formation due to the different zonal movement of the NAO+ and UB dipole anomalies in high and middle latitudes over the North Atlantic and Eurasia. The solid (dashed) lines denote the strong (weak) zonal wind regions that correspond to the eastward (westward) movement of the height anomaly, and the letter H (L) represents the anticyclonic (cyclonic) anomaly.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 148 148 50
PDF Downloads 131 131 39

Ural Blocking as an Amplifier of the Arctic Sea Ice Decline in Winter

View More View Less
  • 1 Institute of Oceanography, Chinese Academy of Science, Qingdao, China
  • 2 CAS Key Laboratory of Regional Climate-Environment for Temperate East Asia, Institute of Atmospheric Physics, Chinese Academy of Science, Beijing, China
© Get Permissions
Full access

Abstract

In this paper, the lead–lag relationship between the Arctic sea ice variability over the Barents–Kara Sea (BKS) and Ural blocking (UB) in winter (DJF) ranging from 1979/80 to 2011/12 is examined. It is found that in a regressed DJF-mean field an increased UB frequency (days) corresponds to an enhanced sea ice decline over the BKS, while the high sea surface temperature over the BKS is accompanied by a significant Arctic sea ice reduction. Lagged daily regression and correlation reveal that the growth and maintenance of the UB that is related to the positive North Atlantic Oscillation (NAO+) through the negative east Atlantic/west Russia (EA/WR) wave train is accompanied by an intensified negative BKS sea ice anomaly, and the BKS sea ice reduction lags the UB pattern by about four days. Because the intensified UB pattern occurs together with enhanced downward infrared radiation (IR) associated with the intensified moisture flux convergence and total column water over the BKS, the UB pattern contributes significantly to the BKS sea ice decrease on a time scale of weeks through intensified positive surface air temperature (SAT) anomalies resulting from enhanced downward IR. It is also found that the BKS sea ice decline can persistently maintain even when the UB has disappeared, thus indicating that the UB pattern is an important amplifier of the BKS sea ice reduction. Moreover, it is demonstrated that the EA/WR wave train formed by the combined NAO+ and UB patterns is closely related to the amplified warming over the BKS through the strengthening (weakening) of mid-to-high-latitude westerly wind in the North Atlantic (Eurasia).

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author e-mail: Dr. Dehai Luo, ldh@mail.iap.ac.cn

Abstract

In this paper, the lead–lag relationship between the Arctic sea ice variability over the Barents–Kara Sea (BKS) and Ural blocking (UB) in winter (DJF) ranging from 1979/80 to 2011/12 is examined. It is found that in a regressed DJF-mean field an increased UB frequency (days) corresponds to an enhanced sea ice decline over the BKS, while the high sea surface temperature over the BKS is accompanied by a significant Arctic sea ice reduction. Lagged daily regression and correlation reveal that the growth and maintenance of the UB that is related to the positive North Atlantic Oscillation (NAO+) through the negative east Atlantic/west Russia (EA/WR) wave train is accompanied by an intensified negative BKS sea ice anomaly, and the BKS sea ice reduction lags the UB pattern by about four days. Because the intensified UB pattern occurs together with enhanced downward infrared radiation (IR) associated with the intensified moisture flux convergence and total column water over the BKS, the UB pattern contributes significantly to the BKS sea ice decrease on a time scale of weeks through intensified positive surface air temperature (SAT) anomalies resulting from enhanced downward IR. It is also found that the BKS sea ice decline can persistently maintain even when the UB has disappeared, thus indicating that the UB pattern is an important amplifier of the BKS sea ice reduction. Moreover, it is demonstrated that the EA/WR wave train formed by the combined NAO+ and UB patterns is closely related to the amplified warming over the BKS through the strengthening (weakening) of mid-to-high-latitude westerly wind in the North Atlantic (Eurasia).

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author e-mail: Dr. Dehai Luo, ldh@mail.iap.ac.cn

1. Introduction

In the past three decades the Arctic sea ice has remarkably declined in all seasons, especially in summer (Wang and Overland 2009; Simmonds 2015). Although the Arctic sea ice decline is weaker in winter than in summer, the summer sea ice variability is sensitive to the sea ice conditions of the previous winter (Rigor et al. 2002; D. S. Park et al. 2015). For this reason, the physical cause of why the Arctic winter sea ice has significantly declined in recent decades has attracted a great attention (Fang and Wallace 1994; Sorteberg and Kvingedal 2006; Francis and Hunter 2006; D. S. Park et al. 2015; H. S. Park et al. 2015; Sorokina et al. 2016).

In recent years, several mechanisms have been proposed to account for the cause of the winter Arctic sea ice decline (Francis et al. 2005; Screen and Simmonds 2010; Screen et al. 2010; Cavalieri and Parkinson 2012). For example, the increased downward infrared radiation (IR) (D. S. Park et al. 2015; H. S. Park et al. 2015), the ice albedo feedback (Lindsay and Zhang 2005; Screen and Simmonds 2010), wind-induced sea ice drift away from the Arctic region through the Fram Strait (Rigor et al. 2002), and the northward transport of atmospheric moisture and sensible heat (Woods et al. 2013; Woods and Caballero 2016; H. S. Park et al. 2015; Burt et al. 2016) have been recognized to be important for the winter Arctic sea ice decline. Although some studies suggested that the winter Arctic sea ice decline is likely associated with changes in atmospheric circulations (Fang and Wallace 1994; Sorteberg and Kvingedal 2006; H. S. Park et al. 2015), it is unknown what certain type of large-scale atmospheric circulation leads to the enhanced variability of the Arctic sea ice in winter. In our previous studies (Luo et al. 2016a,b), we have examined the role of an increased occurrence of Ural blocking (UB) (number of days) in the amplification of the winter warm Arctic–cold Eurasian (WACE) pattern, and conjectured that long-lived UB events can probably lead to the further reduction of the Arctic sea ice over the Barents–Kara Sea (BKS) through the amplification of high-latitude Arctic warming. Unfortunately, no observational evidence is provided to support this hypothesis. To some extent, this hypothesis is inconclusive. At present, the linkage between the Arctic sea ice decline or Arctic warming and midlatitude weather patterns is still a difficult problem (Walsh 2014; Cohen et al. 2014; Overland et al. 2015). In fact, understanding the causal relationship between the UB and Arctic sea ice decrease is a key element of understanding the link between Arctic amplification and midlatitude weather and climate variability. However, identifying the causal relationship between the UB and the sea ice decline over the BKS is difficult because they are often coupled together and cannot be strictly separated. While the winter Arctic warming or the BKS sea ice decrease correspond to a high-value region of the UB frequency based on a regressed winter-mean field (see Fig. 3d) presented below, whether a UB pattern leads to the Arctic sea ice decline over the BKS or vice versa on time scales of weeks is not well understood because the previous results were obtained mainly based on monthly mean sea ice data (Luo et al. 2016a). Maybe this problem can be solved by looking at the daily evolution of the Arctic sea ice and UB using the daily data. In this paper, we will examine the causal relationship between the UB and Arctic sea ice decrease using regression and correlation analyses from a daily perspective.

The most important finding of this paper is that the Arctic sea ice decrease over the BKS lags the UB pattern by about four days, while it can persistently maintain even when the UB has disappeared. This feature cannot be seen if the monthly mean sea ice data are used. Thus, it is thought that the presence of the UB pattern can produce a persistent sea ice decline over the BKS and its adjacent regions in winter, although the UB has a promptly evolved variability with a time scale of about 10–20 days. In this situation, it is concluded that the UB is an amplifier of the Arctic sea ice decline over the BKS region, while the long-time mean BKS winter sea ice reduction related to the high sea surface temperature (SST) provides a condition that favors long-lived UB events through reducing the mid-to-high-latitude westerly wind over Eurasia (Luo et al. 2016a). Such a two-way relationship reflects a positive feedback between the BKS sea ice decline and the UB pattern.

This paper is organized as follows: In section 2, we describe the data and method. In section 3, we present the linear regression maps of atmospheric circulations [DJF-mean 500-hPa geopotential height and surface air temperature (SAT) anomalies] onto the DJF-mean Arctic sea ice time series. Moreover, the linear regression of the DJF-mean Arctic SIC onto the time series of winter SST over the BKS and UB frequency (days) are presented to understand whether or not the contributions of the UB pattern and SST anomalies over the BKS to the Arctic sea ice decrease are equal, although the SST over the Arctic has been suggested to be important for the Arctic sea ice reduction (Francis and Hunter 2007). In section 4, the lead–lag relationship between the UB and sea ice variations is examined using lagged regression analyses from a daily perspective. The physical cause of the negative east Atlantic/west Russia (EA/WR) pattern wave train that the NAO+ and UB combine to form is presented in section 5. Results and discussions are given in section 6.

2. Data and method

Observations have revealed that the Arctic sea ice exhibits a rapid decline in the most recent decade (Francis and Hunter 2007), which actually reflects the combination of a long-term trend and a short-term variability of the Arctic sea ice. As noted in Luo et al. (2016a), there is an increase in UB days in high latitudes in the most recent decade. Thus, it is inferred that the rapid reduction of the Arctic sea ice over the BKS is likely related to the increased UB days if the UB is able to cause the decline of the BKS sea ice on a short time scale of weeks. To examine this problem, our focus is mainly placed on the result of the detrended daily data during 1979–2011 in this study. The daily sea ice concentration (SIC) data for winter [from December to February (DJF)] used are taken from the National Snow and Ice Data Center (NSIDC) dataset, which ranges from 1979/80 to 2011/12 (1979–2011). For the same time range, the atmospheric data, including the 500-hPa geopotential height, SAT, vertical integral of divergence of moisture flux, total column ice and liquid water, downward IR, and the oceanic data, including SST, are derived from the ERA-Interim dataset (Dee et al. 2011), although the SST from the ERA-Interim dataset used here has a small difference with the observations. For the daily data, the anomaly of each variable at each grid point during 1979–2011 is calculated as its deviation from their long-term (1979–2011) mean for each day of the winter. For this case, the atmospheric variables are deseasonalized.

To identify UB events over the Ural Mountain region (40°–80°E), here we use the blocking index of Tibaldi and Molteni (1990) (TM index). The daily TM index is defined based on the southern and northern meridional gradients of daily 500-hPa geopotential height (GHGS and GHGN, respectively) calculated at three latitudes for a given calculation domain: GHGS = Z(ϕ0) − Z(ϕS)/(ϕ0ϕS) and GHGN = Z(ϕN) − Z(ϕ0)/(ϕNϕ0), where Z is the 500-hPa geopotential height with ϕN = 80°N + Δ, ϕ0 = 60°N + Δ, ϕS = 40°N + Δ, and Δ = −5°, 0°, or 5° latitude. A given longitude is defined as “blocked” at a specific instant in time if the following two conditions are satisfied for at least one value of Δ (Tibaldi and Molteni 1990): 1) GHGS > 0 and 2) GHGN < −10 gpm (° lat)−1. A blocking event is defined to have taken place in a given domain if the above two conditions are satisfied simultaneously and required to persist for at least three consecutive days. The persisting days of the simultaneously satisfied two conditions are defined as the blocking duration or the number of blocking days (the blocking days). Based on this definition, the sum of the persisting days of all blocking events occurring over the Ural Mountain region for a winter from December to February can be calculated as the UB frequency in winter when we place our calculation domain in the region (40°–80°E). Like this, it is easy to construct a long time series of the winter UB frequency, as given below. Here, the value of the normalized daily GHGS is defined as the daily TM strength index that is used as a measure of the blocking strength at every day. For this case, the regressed daily field of 500-hPa geopotential height anomaly against the daily TM strength index can reflect the daily evolution of a UB pattern. Moreover, it needs to be pointed out that the TM index as the one-dimensional (1D) index used here is independent of the longitude. But it can become a two-dimensional (2D) blocking index if it depends on the longitude (Diao et al. 2006). The advantage of this 2D blocking index is able to identify the horizontal distribution of blocking days in a wide region (Luo et al. 2016a). However, using a 1D index is a better choice because this method can allow us to not consider the blocking activity outside a given domain. In particular, it can make it easier for us to pick out the number of blocking events and calculate blocking days in the Ural Mountain region.

In this paper, because we have only the daily Arctic sea ice data during 1979–2011, the number of the winter UB days for the same time interval is only calculated to examine the possible connection of the daily variability of the Arctic sea ice to the UB pattern.

3. A connection of the sea ice anomaly over the BKS to the SST and Ural blocking

a. Large-scale circulation patterns associated with the Arctic sea ice decline

To see what types of large-scale circulations and associated SAT anomalies correspond to the Arctic sea ice reduction, we first show the time series of the detrended DJF-mean SIC anomaly over the whole Arctic region (60°–90°N) and the BKS (65°–80°N, 0°–90°E) in Figs. 1a,b. We see that the Arctic sea ice has its peaks in 1987/88 around the Arctic (Fig. 1a), basically consistent with the result of Lindsay and Zhang (2005), who noted that the Arctic sea ice has a thickness peak in 1987 and shows a thinning from 1988. There are still SIC peaks in 1987/88 if the time series of the SIC anomaly averaged over the BKS is only calculated (Fig. 1b), while they are relatively weak compared to those in Fig. 1a. We show the regression fields of DJF-mean 500-hPa geopotential height and SAT anomalies against the time series of the SIC anomalies averaged over the Arctic and the BKS during 1979–2011 (multiplied by −1) in Figs. 1c,d. It is obvious that the Arctic sea ice decline corresponds to positive height anomalies around the Arctic region. The annular-like positive height anomaly looks like a negative-phase Arctic Oscillation, while it is strongest over the BKS near the Ural Mountains (contour in Fig. 1c). The strongest Arctic warming is concentrated in the BKS (color shading in Fig. 1c) as well. A similar result is found if the regressed fields during 1987–2011 are performed, as shown in Fig. 2.

Fig. 1.
Fig. 1.

Time series of DJF-mean SIC anomalies (red dashed line denotes a detrending) over (a) the whole Arctic and (b) BKS (65°–80°N, 0°–90°E) during 1979–2011; and linear regression of the winter-mean 500-hPa geopotential height (contour interval: 50 gpm) and SAT (color shading) anomalies against the detrended DJF-mean SIC index (multiplied by −1) over (c) the whole Arctic and (d) BKS during 1979–2011. The dotted areas show values that are statistically significant at the p < 0.05 level for the Student’s t test.

Citation: Journal of Climate 30, 7; 10.1175/JCLI-D-16-0548.1

Fig. 2.
Fig. 2.

Linear regression of the DJF-mean 500-hPa geopotential height (contour interval: 50 gpm) and SAT (color shading) anomalies against the detrended DJF-mean SIC index (multiplied by −1) over (a) the whole Arctic and (b) BKS during 1987–2011. The dotted areas show values that are statistically significant at the p < 0.05 level for the Student’s t-test.

Citation: Journal of Climate 30, 7; 10.1175/JCLI-D-16-0548.1

If one only considers the time variation of the DJF-mean SIC anomaly averaged over the BKS (Fig. 1b), the regression field of the 500-hPa geopotential height anomaly against the BKS SIC time series (Fig. 1d) shows a dominant positive height anomaly over the Ural Mountain region resembling an UB pattern. This UB pattern is closely related to the positive phase of the NAO (NAO+) (Fig. 1d), which was first noted by Luo (2016a). We also see that, for the regressed DJF-mean 500-hPa height fields against the time SIC series over the whole Arctic (Fig. 2a) and BKS (Fig. 2b) during 1987–2011, there are NAO+ patterns over the North Atlantic, thus indicating that the UB with NAO+ is more pronounced during 1987–2011 than during 1979–2011. The combination of the UB and NAO+ patterns forms a EA/WR wave train pattern (Lim 2015). As noted below, the daily regression height field can also show such a wave train structure. The physical cause of why the NAO+ pattern together with the UB shows an EA/WR pattern will be presented in section 5. In the regressed SAT anomaly field (color shading in Fig. 1d), high-latitude warming and midlatitude cooling can be seen, which correspond just to a warming over the BKS and a cooling over central Asia and reflect the double-sided effect of the UB pattern. Thus, the presence of the UB pattern is able to amplify the Arctic warming over the BKS prior to the block onset (Luo et al. 2016a). In this case, the BKS region experiences an intensified warming, while midlatitude central Asia undergoes a significant cooling. Because this midlatitude cold anomaly (Fig. 1d) resembles the cooling trend pattern (1993–2013) of the DJF-mean SAT over central Asia obtained by Cohen et al. (2014, their Fig. 2c), it is inferred that the cooling trend over central Asia observed in the most recent decade that constitutes a main component of the winter warming hiatus is attributed to an increased frequency of long-lived (persistent) UB patterns after 2000 (Luo et al. 2016a). Although the UB pattern can induce an additional warming over the BKS, how it affects the BKS sea ice variability is not clarified because the monthly mean sea ice data were only used in that study (Luo et al. 2016a). Actually, it is difficult to infer their causal relationship from a winter-mean perspective in that the Arctic sea ice decreases in the BKS region and the associated Arctic warming and UB pattern are strongly coupled together. This motivates us to examine this problem from a daily perspective. Before examining the cause-and-effect relationship between the Arctic sea ice decrease and UB pattern from a daily perspective, it is useful to look at whether the UB pattern contributes to the Arctic sea ice reduction over the BKS from a winter-mean perspective. Of course, it is also useful to make a comparison with the contribution of the wintertime-mean SST anomaly to understand the different roles of the SST anomaly and UB pattern in the BKS sea ice variation (Francis and Hunter 2007), because the Atlantic multidecadal oscillation (AMO) can affect the Arctic sea ice through modulating the SST change over the BKS (Peings and Magnusdottir 2014) or the inflow of North Atlantic warm water into the Arctic region (Alexeev et al. 2013). This comparison can be made by performing the linear regression of the DJF-mean SIC against the time series of the wintertime-mean UB frequency and SST anomaly over the BKS, which is reported in the next subsection.

b. Link of the wintertime mean BKS sea ice decline with the SST and UB pattern on an interannual time scale

We first construct the time series of the DJF-mean SST over the BKS region and UB frequency (days) before performing linear regression. The DJF-mean SST anomaly averaged over the region (65°–80°N, 0°–90°E) is defined to characterize the variation of the winter SST over the BKS, whereas the number of UB days (frequency) per winter is calculated in terms of the definition of the blocking days as given above to construct the time series of the winter-mean UB frequency index during 1979–2011.

The time series of the normalized DJF-mean BKS SST anomaly and UB frequency during 1979–2011 are shown in Figs. 3a,b. It is seen from Fig. 3a that the SST anomaly over the BKS is mostly positive during 2000–11, especially during 2004–11. The main cause of this is that the Atlantic warm water entering the Arctic (i.e., the BKS) is more prominent during 2004–11 than before, probably through the intensified poleward ocean heat transport (Alexeev et al. 2013; Ärthun and Eldevik 2016). The correlation coefficient between the DJF-mean SST and UB frequency time series (Fig. 3b) is 0.39, which is statistically significant at the 95% confidence level, thus indicating that more UB days likely warm the SST in addition to melting sea ice over the BKS, although the high SST favors more UB days through the BKS warming. The coupling between the SST and UB deserves further investigation. But the purpose of this paper is to examine how the UB affects the short-term variability of the sea ice over the BKS.

Fig. 3.
Fig. 3.

Time series of the normalized winter-mean (a) SST over the BKS (the red dashed line denotes a detrending) and (b) UB frequency (days) from 1979–2011 (red dashed line denotes a 5-yr running mean); and linear regression of the DJF-mean (c) SIC against the SST and (d) SIC against the UB frequency during 1979–2011. The color shading denotes the region above the 90% confidence level for the Student’s t test.

Citation: Journal of Climate 30, 7; 10.1175/JCLI-D-16-0548.1

Although the main feature of the normalized DJF-mean UB frequency time series (Fig. 3b) is consistent with that of Luo et al. (2016a), there is a small difference of the UB frequency time series with the result of Luo et al. (2016a) because of the time period interval and its definition being slightly different. In Luo et al. (2016a), the time interval from December 1978 to February 2014 is defined as the period 1979–2013. Here, the time interval from December 1979 to February 2012 is defined as 1979–2011. Such a definition difference gives rise to a difference between the two curves of the DJF-mean UB frequency time series obtained for 1979–2011 and 1979–2013. However, we must emphasize that the definition of the time period presented here seems to be more reasonable. Although the upward trend of the wintertime UB frequency during 2000–11 in Fig. 3b is not so obvious, it can become more significant once the UB events of 2012/13 and 2014/15 are included in the UB frequency time series in Fig. 3b (Luo et al. 2016a). However, it is also useful to examine whether the BKS sea ice decline is related to the UB pattern from a winter-mean perspective, while our purpose is to focus on examining the daily relationship between the SIC decline and UB pattern.

We show regressed DJF-mean SIC anomaly fields against the time series of the DJF-mean SST (Fig. 3a) and UB frequency (Fig. 3b) during 1979–2011 in Figs. 3c and 3d, respectively. It is seen from the regressed field against the BKS SST time series that the SIC anomaly shows a significant decrease over the BKS and its adjacent sea region east of Greenland (blue color in Fig. 3c). A similar winter SIC variability pattern is also seen for the regressed SIC anomaly field against the UB frequency time series (blue color in Fig. 3d), even though it is weaker than the SST-related SIC anomaly. This implies that the trend of Arctic winter sea ice decline over the BKS is not only associated with the SST over the BKS, but also with the UB frequency, while the high SST-related sea ice decline is more prominent than the UB-related sea ice decrease. As discussed in the final section, the high SST over the BKS and its influence on the Arctic sea ice are more likely modulated by the phase of the AMO. The study of this problem is beyond the scope of the present paper. Although the UB pattern corresponds to the Arctic sea ice decline over the BKS (Fig. 3d), it is unclear whether the UB leads to the BKS sea ice decrease or vice versa. This can be investigated by using the daily data. In previous studies, many people have examined the physical cause of the trend of Arctic sea ice decline (Francis and Hunter 2007; Peings and Magnusdottir 2014; H. S. Park et al. 2015), but the causal relationship between the atmospheric circulation changes and Arctic sea ice decline is still unclear (Fang and Wallace 1994). In this paper, our intention is to reveal the cause-and-effect relationship between the BKS sea ice variability and UB pattern on a time scale of weeks, even though the long-term variation of the UB frequency can probably affect the trend of Arctic sea ice decline over the BKS. As noted by Luo et al. (2016a), the long-time mean Arctic sea ice decline favors the UB pattern with a time scale of two weeks through the weakening of mid-to-high-latitude westerly wind over Eurasia due to the Arctic warming. But as we will see below, the presence of the UB pattern can induce a positive SAT anomaly with a time scale of two weeks over the BKS to produce the Arctic sea ice decline on a time scale of weeks, thus strengthening the Arctic sea ice reduction prior to the block onset. In this case, the UB may be thought of as being an amplifier of the Arctic sea ice decline. Of course, the UB-related Arctic sea ice decrease will have a long-term trend if the UB frequency shows a long-term trend. This problem remains to be further examined in another paper. Moreover, another point different from Fig. 3c is that a significant decrease in the Arctic sea ice is seen over the Bering Sea and its Arctic region around 180° longitude (Fig. 3d) because a blocking anomaly exists also over its east side.

The correlation calculation shows that the Arctic winter sea ice reduction is indeed associated with the increased UB frequency, because the time series of the detrended normalized DJF-mean SIC and UB frequency show a negative correlation of −0.42, which is statistically significant at the 95% confidence level. However, because the UB and the Arctic sea ice decrease over the BKS are coupled together through the Arctic warming, it is difficult to discern whether the UB drives the BKS sea ice decline through Arctic warming or whether the BKS sea ice reduction drives the UB pattern through warming overlying atmosphere from a DJF-mean perspective. Although Luo et al. (2016a) surmised that the long-lived UB is likely to cause the BKS sea ice decline, it is probably because of the DJF-mean high-latitude warming over the BKS amplified by the UB pattern. But such a causal relationship is unclear and not quantified from a daily perspective. Luo et al. (2016b) also noted that the UB pattern does not arise directly from the Arctic warming related to Arctic Sea ice decrease, while the Arctic warming provides a favorable condition that weakens mid-to-high-latitude westerly wind to favor the incidence of blocking occurrence. Instead, the UB results mainly from the low-frequency wave train propagation or wave energy dispersion originating from the North Atlantic through the breakdown of the NAO+ (Luo et al. 2007). Thus, it is hypothesized that the UB pattern has a resulting effect that gives rise to a significant reduction of the sea ice over the BKS region because it is often accompanied by a high-latitude warming. To testify this hypothesis, it is necessary to examine the lead–lag relationship between the UB and SIC variability using the daily data in the next subsection.

c. Relationship between the daily UB and SIC decrease on a time scale of weeks

Here, we try to infer causality between the UB and SIC through examining their daily evolution and lead–lag relationship. We show the 2-day-interval lagged regression fields of the daily 500-hPa geopotential height anomalies, SAT, and SIC against the daily TM strength index during the 1979–2011 winters in Fig. 4. It is seen that a positive height anomaly as a weak UB is very weak over the Ural Mountain region on lag −10 days (Fig. 4a) and then intensifies and becomes quasi stationary. There is a clear large-scale negative-over-positive dipole anomaly over the North Atlantic at lag −8 days, which resembles an NAO+ pattern that evolves with the intensification of the UB anomaly. Then this NAO+ pattern is gradually intensified until lag −4 days. Along with the decay of the NAO+ pattern after lag −4 days, this blocking anticyclone is intensified to form a typical quasi-stationary UB pattern from lag −8 to 0 days and reaches its largest amplitude on lag 0 days. Afterward, it decays till lag +8 days. We also see that in a daily regressed height field the NAO+ and UB patterns combine to form a Rossby wave train structure similar to an EA/WR pattern noted above. Thus, it is concluded from these results that the NAO+ pattern can drive the generation of the UB pattern due to its downstream energy dispersion through the EA/WR wave train propagation (Luo et al. 2007). This NAO+ pattern has been shown to precede the UB by 4–7 days (Luo et al. 2016b). In particular, this NAO+ pattern is almost invisible after lag +4 days (Fig. 4a). The energy of the UB pattern comes mainly from the decay of the NAO+ pattern (Luo et al. 2016b). Moreover, it is seen from Fig. 4b that the positive SAT anomaly over the BKS is weak during the period from lag −10 to −6 days because of the UB amplitude being small. Along with the intensification of the UB, a strong positive SAT anomaly appears over the BKS after lag −4 days and is further intensified and expanded toward the continental side of Eurasia, and then it begins to slowly decline after it reaches its strongest amplitude at lag 0 days. While this positive SAT anomaly is weak after lag 6 days, it can still persist (Fig. 4b for lag +10 days), even when the UB amplitude is small (Fig. 4a). This indicates that the persistent positive SAT anomaly over the BKS is associated with the establishment and maintenance of the UB pattern.

Fig. 4.
Fig. 4.

Lagged regressions of the daily (a) 500-hPa geopotential height, (b) SAT, and (c) SIC anomalies against the daily TM strength index for the 1979–2011 winters (DJF). The color shading denotes the regions above the 95% confidence level for the two-sided Student’s t test. Lag 0 denotes the day of the UB peak.

Citation: Journal of Climate 30, 7; 10.1175/JCLI-D-16-0548.1

Figure 4c reveals that, while the SIC anomaly is mostly positive over the BKS from lag −10 to −6 days, the positive SIC anomaly weakens along with the growth of the UB pattern. A negative SIC anomaly over the BKS region is seen to appear on lag −4 days and becomes strong after this day. An interesting result we see is that the strong SIC decline (large negative SIC anomaly) with a short time scale of weeks can persist, even though the UB is very weak and has disappeared after lag 8 days (Fig. 4a). In particular, a strong sea ice decline over the BKS can also be seen during the period from lag 10 to 12 days (Fig. 4c). This reflects that the Arctic sea ice decline is a slow process compared to the UB decay. Because the SIC decline reaches a maximum value at about lag +4 days (see Fig. 6), the SIC decline lags the UB peak by about 4 days. Thus, it is deduced that the occurrence of the UB pattern is able to lead to the BKS sea ice decline through the persistent positive SAT anomaly over the BKS because the SAT peak leads the peak of the BKS sea ice decline. The above results are also held for the 1987–2011 winters (not shown). However, the precise lead–lag relationship between the UB and the sea ice decrease over the BKS needs to be further quantified by calculating the pattern correlations between the lagged regressions of the atmospheric variables (daily 500-hPa geopotential height, SAT, and SIC) and the UB pattern and by examining the time evolution of associated variables (downward IR, moisture, and total column water).

Some studies have indicated that the enhanced downward IR associated with the intensified moisture flux convergence from midlatitudes into the Arctic is important for the reduction of Arctic sea ice (D. S. Park et al. 2015; Woods and Caballero 2016). Below, we will indicate that changes in the moisture flux convergence, total column water (TCW; liquid water plus ice), and associated downward IR over the BKS depend strongly on the evolution (growth and decay) of the UB pattern from a daily perspective. In particular, it is demonstrated that the intensified UB occurs together with enhanced positive SAT anomaly, downward IR, TCW, and moisture flux convergence over the BKS and its adjacent region that drive the reduction of the Arctic sea ice over the BKS.

4. A link of the daily variations of moisture flux convergence, total column water, and downward IR with the evolution of the UB pattern

As demonstrated by many previous studies (Ghatak and Miller 2013; D. S. Park et al. 2015; H. S. Park et al. 2015), the Arctic winter sea ice reduction is associated with the positive SAT anomaly over the Arctic that is mainly driven by the intensified downward IR. In fact, the variability of the downward IR is closely related to changes in moisture flux convergence and TCW over the Arctic, both being excellent emitters of downward IR. Thus, it is useful to show linear regression of the moisture flux convergence (multiplied by the latent heat of vaporization, L = 2.26 × 106 m2 s−2), TCW, and downward IR against the daily TM strength index during 1979–2011 in Fig. 5.

Fig. 5.
Fig. 5.

Lagged regressions of the daily (a) vertically integrated moisture flux convergence (multiplied by L), (b) TCW, and (c) downward IR anomalies against the daily TM strength index for the 1979–2011 winters. The color shading in (b) and (c) denotes the region above the 95% confidence level for the two-sided Student’s t test, whereas the color shading in (a) represents the region above the 90% confidence level. Lag 0 denotes the day of the UB peak.

Citation: Journal of Climate 30, 7; 10.1175/JCLI-D-16-0548.1

It is found from Fig. 5a that the moisture flux convergence anomalies are weak over the BKS at lag −10 and −8 days. Instead, there is a relatively strong moisture flux convergence region in the southeast side of Greenland. The moisture flux convergence from the North Atlantic is intensified and widened and then extends to the BKS region with the growth of the UB pattern (from lag −8 to 0 days in Fig. 5a). But it then weakens so rapidly (from lag 0 to 4 days in Fig. 5a) that only weak moisture flux convergence can be seen over the BKS region after lag +4 days once the UB decays (Fig. 5a). A comparison of Fig. 5a with Fig. 4a also reveals that the intensified moisture flux convergence over the BKS follows the occurrence and intensification of the UB pattern. Because the southerly wind prevails in the upstream side of the UB anticyclone, the intensified southerly wind upstream of the UB pattern can bring more North Atlantic warm moist air from midlatitudes to the Arctic mainly over the BKS region (from lag −6 to 0 days in Fig. 5a) through the Greenland Sea and Norwegian Sea as the UB strengthens. Meanwhile, we see that the TCW anomalies over the BKS are almost invisible during the beginning phase of the UB pattern (from lag −10 to −8 days in Fig. 5b). Growing TCW is found after lag −6 days (from lag −6 to 0 days in Fig. 5b) as the UB approaches its maximum amplitude (Fig. 4a). Then the TCW is decreased with the decay of the UB pattern, as shown in Fig. 5b from lag 0 to 6 days. Thus, the increased TCW over the BKS is closely related to the intensification and maintenance of the UB pattern. Figure 5c shows the daily regressed downward IR anomalies against the daily TM strength index during 1979–2011. It is noted that the downward IR is weakly negative over the BKS region during the period from lag −10 to −8 days but becomes positive, strengthens, and expands toward the midlatitude region after lag −8 days. Thus, it is obvious that such an increased positive downward IR anomaly occurs together with the intensification of the UB pattern (from lag −8 to 0 days in Fig. 5c). It can be persistent, even though it decreases slightly (from lag 0 to +8 days in Fig. 5c) as the UB pattern decays. Thus, it is suggested that the enhanced downward IR is mostly due to the increased moisture flux convergence and TCW, both associated with the intensified UB pattern.

Moreover, a comparison with Fig. 4c shows that a large increase in the downward IR from lag −2 to 4 days in Fig. 5c precedes a large decrease of the Arctic sea ice in the region from the east of Greenland to the BKS from lag −2 to 12 days (Fig. 4c). Therefore, along with the further intensification of the downward IR, a significant decline of the Arctic sea ice over the BKS is seen as a result of the intensified downward IR. Meanwhile, because the Arctic sea ice reduction is a slow process, it can persist (Fig. 4c), even though the downward IR is weak in the region from the east of Greenland to the BKS region during the period from lag +10 to 12 days (Fig. 5c).

The above results reveal that the changes in the moisture flux convergence, TCW, and downward IR are closely associated with the establishment and maintenance of the UB pattern. The Arctic winter sea ice decline can be attributed to the intensified moisture flux convergence, TCW, and associated intensified downward IR over the BKS once the UB pattern is generated and maintained. In the next subsection, we will further quantify their lead–lag relationship with the SIC, SAT, and UB, using the daily pattern correlations.

Lead–lag relationship of the moisture flux convergence, total column water, downward IR, and UB with the SIC decrease

We show the pattern correlation coefficients of 500-hPa geopotential height, moisture flux convergence, TCW, downward IR, and SAT anomalies and the same variables obtained by regressing against the daily TM strength index (lag 0 days as the UB peak in Fig. 4a) for the 1979–2011 winters in Fig. 6. The lead–lag relationship between the UB and the sea ice decline can also be examined by calculating the pattern correlation between the lagged regressions of the daily SIC and the same variable obtained by regressing against the winter UB frequency (Fig. 3b) over a range of time lags, as shown in Fig. 6. It is interesting to see that the pattern correlation coefficients can reach one at lag 0 days because these variables can have their peaks in the same time (Figs. 4 and 5), while the pattern correlation coefficient of the SIC anomaly cannot reach one because the SIC decline (its peak is not at lag 0 days in Fig. 4c) has a time lag difference with the UB. The above result shows that there is no time lag between any one of the aforementioned atmospheric variables (500-hPa geopotential height, moisture flux convergence, TCW, downward IR, and SAT anomalies) and the UB, thus indicating that the intensified moisture flux convergence (Fig. 5a), TCW (Fig. 5b), downward IR (Fig. 5c), and SAT (Fig. 4b) exhibit consistent variations with the intensified UB pattern (Fig. 4a). But a large time lag between the SIC anomaly and UB can be seen in Fig. 6, in which the maximum SIC decline takes place at about lag +4 days relative to the UB peak (lag 0 day). Thus, the atmospheric variables (the moisture flux convergence, TCW, downward IR, SAT, and the UB pattern) lead the peak of the SIC decrease by about 4 days. Therefore, it is suggested that the UB can give rise to a significant decline of the BKS sea ice through the intensified SAT over the BKS as a result of enhanced downward IR associated with the intensified moisture flux convergence and TCW. On the other hand, an interesting result is found that the sea ice decrease over the BKS region continues to maintain its large amplitude (red curve in Fig. 6) even when the UB has disappeared during the period from lag 10 to 20 days. Thus, it is deduced that the UB can induce a persistent reduction of the Arctic sea ice over the BKS region (Fig. 4c and red curve in Fig. 6) through a persistent positive SAT anomaly (purple curve in Fig. 6) associated with the persistent strong downward IR (green curve in Fig. 6). This result is a new finding not reported by previous studies (Francis and Hunter 2007; D. S. Park et al. 2015; H. S. Park et al. 2015).

Fig. 6.
Fig. 6.

Pattern correlations between the daily 500-hPa geopotential height (Hgt_500mb), moisture flux convergence (Mois_con), TCW (liquid water plus ice), and downward IR and SAT anomalies and the same variables obtained by lagged regression against the daily TM strength index for the 1979–2011 winters. The sea ice pattern correlation (red line) is derived by performing the lagged regressions of the daily SIC and the same variable obtained by regressing against the wintertime UB frequency (see Fig. 3b). Lag 0 denotes the day of the UB peak. The dots show values that are statistically significant at the p < 0.05 level for the Student’s t test.

Citation: Journal of Climate 30, 7; 10.1175/JCLI-D-16-0548.1

5. Physical cause of the EA/WR wave train structure associated with the NAO+ and UB patterns

While the UB that is regarded as one part of the EA/WR wave train is seen to occur often together with the decay of the NAO+ based on a nonlinear multiscale interaction model (Luo et al. 2007), it is useful to present a physical explanation on the formation of the EA/WR wave train structure. Although Luo et al. (2016b) concluded that the generation of this EA/WR wave train is related to changes in the meridional position and strength of the North Atlantic jet, why the NAO+ and UB patterns can form a wave train structure is not clarified theoretically. Here, we conclude that the generation of this wave train is closely related to the amplified Arctic warming over the BKS. This point was not mentioned in Luo et al. (2016b). To indicate this point, we show the time series of the SAT anomaly averaged over the BKS in Fig. 7a as a measure of the Arctic warming over the BKS. Correspondingly, the regressed 300-hPa zonal wind fields against the time series of the SAT (Fig. 7a) and SST (Fig. 3a) anomalies averaged over the BKS are shown in Figs. 7b,c. It is easy to find that the zonal wind over the North Atlantic depends strongly on the Arctic warming over the BKS (Fig. 7b) rather than on the SST anomaly in the BKS region (Fig. 7c). It is also seen that the Arctic warming in the BKS region can correspond to a strong positive-over-negative westerly wind anomaly over the North Atlantic mid-to-high latitudes, while there is a weak negative westerly wind anomaly over northern Greenland and its Arctic side. For this case, a strong negative-over-positive westerly wind anomaly is seen over Eurasia as well (Fig. 7b). Such dipole zonal wind anomaly distributions can lead to the southwest–northeast (SW–NE) [northwest–southeast (NW–SE)] tilting of the dipole height anomaly over the North Atlantic (Eurasia) because the north center of the dipole height anomaly undergoes an eastward (westward) displacement over the high-latitude North Atlantic (Eurasia), as shown in Fig. 8, while the reversed displacement is seen in midlatitude regions (Fig. 8). Thus, the different spatial tilting of dipole height anomalies over the North Atlantic and Eurasia can form a wave train structure similar to an EA/WR+ pattern (Fig. 8). Such a wave train tends to be quasi stationary through the combination of the eastward- and westward-displaced northern centers of the NAO+ and UB anomalies and allows a quasi-stationary UB to occur. Moreover, it needs to be pointed out that, because the change in the zonal wind anomaly is not strongly sensitive to the SST variation over the BKS (Fig. 7c), the influence of the high SST over the BKS on the generation of the EA/WR pattern is likely unimportant or secondary compared to that of the BKS warming (Fig. 7b). In other words, the EA/WR pattern from the North Atlantic to Eurasia is mainly attributed to the amplified Arctic warming over the BKS, while the UB results from the decay of the NAO+ pattern (Luo et al. 2007, 2016b).

Fig. 7.
Fig. 7.

(a) Time series of the SAT anomaly averaged over the BKS (red dashed curve denotes a detrending) during 1979–2011 and regressed 300-hPa zonal wind fields against the detrended (b) SAT in (a) and (c) SST anomalies averaged over the BKS in Fig. 3a. The dotted areas show values that are statistically significant at the p < 0.05 level for the two-sided Student’s t test.

Citation: Journal of Climate 30, 7; 10.1175/JCLI-D-16-0548.1

Fig. 8.
Fig. 8.

Schematic diagram of the EA/WR pattern formation due to the different zonal movement of the NAO+ and UB dipole anomalies in high and middle latitudes over the North Atlantic and Eurasia. The solid (dashed) lines denote the strong (weak) zonal wind regions that correspond to the eastward (westward) movement of the height anomaly, and the letter H (L) represents the anticyclonic (cyclonic) anomaly.

Citation: Journal of Climate 30, 7; 10.1175/JCLI-D-16-0548.1

6. Results and discussion

In this paper, we have examined the causal relationship between the UB pattern and Arctic sea ice variability over the BKS from a daily perspective. It is found that the UB pattern is an important factor for the variability of the BKS sea ice on a time scale of weeks. Lagged daily regression fields reveal that the moisture flux convergence, total column ice and liquid water (TCW), downward IR, and SAT anomalies over the BKS show an in-phase variation with that of the UB pattern that are intensified (weakened) along with the growth (decay) of the UB, while the variation of the Arctic sea ice over the BKS lags changes in these variables. The main results obtained in this paper are described below:

  1. On a short time scale of weeks, the BKS sea ice decline lags the UB peak by about four days, thus indicating that the decline of the Arctic sea ice over the BKS is a delayed response to the UB pattern that can be persistently maintained even when the UB has disappeared. In this case, the UB pattern can be thought of as being an important amplifier of the Arctic sea ice decline, even though the long-time mean Arctic sea ice reduction over the BKS provides a background that favors increased UB days, as noted in Luo et al. (2016a). This reflects a positive feedback of the UB pattern on the Arctic sea ice decrease.
  2. During the life cycle of the UB, the BKS sea ice decline can be attributed to the intensified positive SAT anomaly over the BKS resulting from intensified downward IR that is associated with enhanced moisture flux convergence and TCW that occur together with the intensified UB pattern.
  3. The UB pattern is seen to occur together with the evolution of the NAO+ pattern through the EA/WR wave train propagation. The formation of the EA/WR wave train structure and the associated quasi-stationary UB are closely related to the amplified warming over the BKS (Fig. 7a) since the BKS warming can lead to a strong positive-over-negative dipole zonal wind anomaly over the North Atlantic and a reversed dipole zonal wind anomaly over Eurasia (Fig. 7b), inducing the SW–NE (NW–SE) tilting of the NAO+ (UB) anomaly (Fig. 8) that promotes quasi-stationary planetary wave train propagation.

Although we have examined the causal relationship between the UB and Arctic sea ice variability on a time scale of weeks, how the long-time variation of the UB frequency affects the trend of Arctic sea ice decline is not investigated in this paper. Further exploration of this question will help us improve our understanding of the physical cause of the rapid decline of the Arctic sea ice observed in the most recent decade, although it is beyond the scope of this paper. Because the phase of the Atlantic multidecadal oscillation (AMO) can modulate the SST and SAT anomalies over the BKS and lead the NAO by 10–15 yr (Chylek et al. 2009; Peings and Magnusdottir 2014), the long-term trend of the BKS sea ice decline is more likely related to the AMO through the changes in the SST and SAT anomalies over the BKS. Thus, it is concluded that both the UB and NAO+ patterns via the EA/WR wave train propagation and associated Arctic sea ice variation are likely modulated by the phase of the AMO (Miles et al. 2014) through the Arctic SST and SAT changes (Chylek et al. 2009). However, the physical mechanisms of how the AMO affects the long-term trend of the Arctic sea ice and what different roles the AMO and UB play in the Arctic sea ice variability are still unclear. In particular, whether the UB pattern shows a multidecadal variation due to the modulation of the AMO and whether it can have an important effect on the Arctic sea ice decrease on longer time scales is not clarified as well. These questions remain to be investigated further.

Acknowledgments

The authors acknowledge support from the National Science Foundation of China (41305048, 41430533, and 41375067) and the National Science and Technology Major Project of China (2016YFA0601802). The authors thank three anonymous reviewers for their useful suggestions in improving this manuscript. Drs. Steven Feldstein and Sukyoung Lee are highly appreciated for their useful discussions on this manuscript when the authors were visiting The Pennsylvania State University.

REFERENCES

  • Alexeev, V. A., V. V. Ivanov, R. Kwok, and L. H. Smedsrud, 2013: North Atlantic warming and declining volume of arctic sea ice. Cryosphere Discuss., 7, 245265, doi:10.5194/tcd-7-245-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ärthun, M., and T. Eldevik, 2016: On anomalous ocean heat transport toward the Arctic and associated climate predictability. J. Climate, 29, 689704, doi:10.1175/JCLI-D-15-0448.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Burt, M. A., D. A. Randall, and M. D. Branson, 2016: Dark warming. J. Climate, 29, 705719, doi:10.1175/JCLI-D-15-0147.1.

  • Cavalieri, D. J., and C. L. Parkinson, 2012: Arctic sea ice variability and trends, 1979–2010. Cryosphere, 6, 881889, doi:10.5194/tc-6-881-2012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chylek, P., C. K. Folland, G. Lesins, M. K. Dubey, and M. Wang, 2009: Arctic air temperature change amplification and the Atlantic Multidecadal Oscillation. Geophys. Res. Lett., 36, L14801, doi:10.1029/2009GL038777.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cohen, J., and Coauthors, 2014: Recent Arctic amplification and extreme mid-latitude weather. Nat. Geosci., 7, 627637, doi:10.1038/ngeo2234.

  • Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Diao, Y., J. Li, and D. Luo, 2006: A new blocking index and its application: Blocking action in the Northern Hemisphere. J. Climate, 19, 48194839, doi:10.1175/JCLI3886.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Fang, Z., and J. M. Wallace, 1994: Arctic sea ice variability on a timescale of weeks and its relation to atmospheric forcing. J. Climate, 7, 18971914, doi:10.1175/1520-0442(1994)007<1897:ASIVOA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francis, J. A., and E. Hunter, 2006: New insight into the disappearing Arctic sea ice. Eos, Trans. Amer. Geophys. Union, 87, 509524, doi:10.1029/2006EO460001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francis, J. A., and E. Hunter, 2007: Drivers of declining sea ice in the Arctic winter: A tale of two seas. Geophys. Res. Lett., 34, L17503, doi:10.1029/2007GL030995.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francis, J. A., E. Hunter, J. R. Key, and X. Wang, 2005: Clues to variability in Arctic minimum sea ice extent. Geophys. Res. Lett., 32, L21501, doi:10.1029/2005GL024376.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ghatak, D., and J. Miller, 2013: Implications for Arctic amplification of changes in the strength of the water vapor feedback. J. Geophys. Res. Atmos., 118, 75697578, doi:10.1002/jgrd.50578.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lim, Y. K., 2015: The east Atlantic/west Russia (EA/WR) teleconnection in the North Atlantic: Climate impact and relation to Rossby wave propagation. Climate Dyn., 44, 32113222, doi:10.1007/s00382-014-2381-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lindsay, R. W., and J. Zhang, 2005: The thinning of Arctic sea ice, 1988–2003: Have we passed a tipping point? J. Climate, 18, 48794894, doi:10.1175/JCLI3587.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Luo, D., A. Lupo, and H. Wan, 2007: Dynamics of eddy-driven low-frequency dipole modes. Part I: A simple model of North Atlantic Oscillations. J. Atmos. Sci., 64, 338, doi:10.1175/JAS3818.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Luo, D., Y. Xiao, Y. Yao, A. Dai, I. Simmonds, and C. Franzke, 2016a: Impact of Ural blocking on winter warm Arctic–cold Eurasian anomalies. Part I: Blocking-induced amplification. J. Climate, 29, 39253947, doi:10.1175/JCLI-D-15-0611.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Luo, D., Y. Xiao, Y. Diao, A. Dai, C. Franzke, and I. Simmonds, 2016b: Impact of Ural blocking on winter warm Arctic–cold Eurasian anomalies. Part II: The link to the North Atlantic Oscillation. J. Climate, 29, 39493971, doi:10.1175/JCLI-D-15-0612.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Miles, M. W., D. V. Divine, T. Furevik, E. Jansen, M. Moros, and A. E. J. Ogilvie, 2014: A signal of persistent Atlantic multidecadal variability in Arctic sea ice. Geophys. Res. Lett., 41, 463469, doi:10.1002/2013GL058084.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Overland, J., J. Francis, R. Hall, E. Hanna, S. Kim, and T. Vihma, 2015: The melting Arctic and midlatitude weather patterns: Are they connected? J. Climate, 28, 79177932, doi:10.1175/JCLI-D-14-00822.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Park, D.-S., S. Lee, and S. B. Feldstein, 2015: Attribution of the recent winter sea-ice decline over the Atlantic sector of the Arctic Ocean. J. Climate, 28, 40274033, doi:10.1175/JCLI-D-15-0042.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Park, H. S., S. Lee, S. W. Son, S. B. Feldstein, and Y. Kosaka, 2015: The impact of poleward moisture and sensible heat flux on Arctic winter sea ice variability. J. Climate, 28, 50305039, doi:10.1175/JCLI-D-15-0074.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Peings, Y., and G. Magnusdottir, 2014: Forcing of the wintertime atmospheric circulation by the multidecadal fluctuations of the North Atlantic Ocean. Environ. Res. Lett., 9, 034018, doi:10.1088/1748-9326/9/3/034018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rigor, I. G., J. M. Wallace, and R. L. Colony, 2002: Response of sea ice to the Arctic Oscillation. J. Climate, 15, 26482663, doi:10.1175/1520-0442(2002)015<2648:ROSITT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Screen, J. A., and I. Simmonds, 2010: The central role of diminishing sea ice in recent Arctic temperature amplification. Nature, 464, 13341337, doi:10.1038/nature09051.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Screen, J. A., and Coauthors, 2010: Increasing fall–winter energy loss from the Arctic Ocean and its role in Arctic temperature amplification. Geophys. Res. Lett., 37, L16707, doi:10.1029/2010GL044136.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Simmonds, I., 2015: Comparing and contrasting the behaviour of Arctic and Antarctic sea ice over the 35 year period 1979–2013. Ann. Glaciol., 56, 1828, doi:10.3189/2015AoG69A909.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sorokina, S. A., C. Li, J. J. Wettstein, and N. G. Kvamstø, 2016: Observed atmospheric coupling between Barents Sea ice and the warm-Arctic cold-Siberian anomaly pattern. J. Climate, 29, 495511, doi:10.1175/JCLI-D-15-0046.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sorteberg, A., and B. Kvingedal, 2006: Atmospheric forcing on the Barents Sea winter ice extent. J. Climate, 19, 47724784, doi:10.1175/JCLI3885.1.

  • Tibaldi, S., and F. Molteni, 1990: On the operational predictability of blocking. Tellus, 42A, 343365, doi:10.1034/j.1600-0870.1990.t01-2-00003.x.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Walsh, J. E., 2014: Intensified warming of the Arctic: Causes and impacts on middle latitudes. Global Planet. Change, 117, 5263, doi:10.1016/j.gloplacha.2014.03.003.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wang, M., and J. E. Overland, 2009: A sea ice free summer Arctic within 30 years? Geophys. Res. Lett., 36, L07502, doi:10.1029/2009GL037820.

  • Woods, C., and R. Caballero, 2016: The role of moist intrusions in winter Arctic warming and sea ice decline. J. Climate, 29, 44734485, doi:10.1175/JCLI-D-15-0773.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Woods, C., R. Caballero, and G. Svensson, 2013: Large-scale circulation associated with moisture intrusions into the Arctic during winter. Geophys. Res. Lett., 40, 47174721, doi:10.1002/grl.50912.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save