Dynamic Linkage Between Cold Air Outbreaks and Intensity Variations of the Meridional Mass Circulation

3214 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

Dynamic Linkage between Cold Air Outbreaks and Intensity Variations of the Meridional Mass Circulation

YUEYUE YU State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, and University of Chinese Academy of Sciences, Beijing, China

RONGCAI REN State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, and Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, and Key Laboratory of Meteorological Disaster (KLME), Nanjing University of Information Science and Technology, Nanjing, China

MING CAI Department of Earth, Ocean and Atmospheric Science, Florida State University, Tallahassee, Florida

(Manuscript received 29 December 2014, in ﬁnal form 24 April 2015)

ABSTRACT

This study investigates the dynamical linkage between the meridional mass circulation and cold air outbreaks using the ERA-Interim data covering the period 1979–2011. It is found that the onset date of continental-scale cold air outbreaks coincides well with the peak time of stronger meridional mass circulation 2 events, when the net mass transport across 608N in the warm or cold air branch exceeds ;88 3 109 kg s 1. 2 During weaker mass circulation events when the net mass transport across 608N is below ;71.6 3 109 kg s 1, most areas of the midlatitudes are generally in mild conditions except the northern part of western Europe. Composite patterns of circulation anomalies during stronger mass circulation events greatly resemble that of the winter mean, with the two main routes of anomalous cold air outbreaks being along the climatological routes of polar cold air: namely, via East Asia and North America. The Siberian high shifts westward during stronger mass circulation events, opening up a third route of cold air outbreaks through eastern Europe, where lies the poleward warm air route in the winter-mean condition. The strengthening of the Icelandic low and Azores high during stronger mass circulation events acts to close off the climatological-mean cold air route via western Europe; this is responsible for the comparatively normal temperature there. The composite pattern for weaker mass circulation events is generally reversed, where the weakening of the Icelandic low and Azores high, corresponding to the negative phase of the North Atlantic Oscillation (NAO), leads to the reopening and strengthening of the equatorward cold air route through western Europe, which is responsible for the cold anomalies there.

1. Introduction Konrad 1996). For instance, cold air outbreaks over North America have been found to be closely related Various anomalous synoptic- and planetary-scale to positive sea level pressure anomalies along the patterns have been identiﬁed as precursor signals for Alaska–Yukon border (Walsh et al. 2001)ortheco- cold air outbreaks over different regions (e.g., Wexler existence of a ridge over the Arctic and a trough over 1951; Colucci and Davenport 1987; Walsh et al. 2001; the Great Lakes region (Konrad 1996). Elsewhere, in East Asia, the intensiﬁcation and expansion of the Siberian high are known to be the triggering mecha- Corresponding author address: Dr. Rongcai Ren, LASG, In- stitute of Atmospheric Physics, CAS, P.O. Box 9804, Beijing nism for cold air surges (Ding 1990; Zhang et al. 1997; 100029, China. Gong and Ho 2004; Takaya and Nakamura 2005). E-mail: [email protected] Palmer (2014) pointed out that intensifying Rossby

DOI: 10.1175/JAS-D-14-0390.1

Ó 2015 American Meteorological Society Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3215 waves within the jet stream, excited by the latent heat the tropical heating source to the polar heating sink via a release over the warming tropical west Pacific may poleward warm air branch in the upper troposphere have contributed to the extremely cold 2013–14 winter (and also the stratosphere in the winter hemisphere) and in the United States. an equatorward cold branch in the lower troposphere.1 There is also ample evidence indicating a robust re- Cold air outbreaks can be directly related to an anom- lationship between continental-scale cold air outbreaks alous strong meridional mass circulation, with more cold and the leading oscillation modes in the winter extra- air discharged from the northern polar region into tropics. It is well recognized that the negative phase of the lower latitudes within the cold air branch. The the North Atlantic Oscillation (NAO) coincides with strengthening of the cold air branch is connected with cold anomalies over Europe and warming in the north- the strengthening of the warm air branch in the upper west Atlantic (Rogers and van Loon 1979; Hurrell and atmosphere and is driven by the amplification of large- van Loon 1997). The studies by Walsh et al. (2001) and scale waves in the midlatitudes. Therefore, the meridi- Cellitti et al. (2006) show that the NAO tends to be in its onal mass circulation perspective not only allows us to negative phase 3–6 days prior to cold air outbreaks over capture the preferred routes of cold air outbreaks di- different regions of the United States and Europe. Luo rectly but also can help us to investigate the precursory et al. (2014) related the cold air outbreak event in changes in various circulation fields for cold air January–February 2012 to a positive-to-negative phase outbreaks. transition of NAO. As a broader anomaly pattern than Iwasaki and Mochizuki (2012) and Iwasaki et al. the NAO, the Arctic Oscillation (AO), or the tropo- (2014) identified two major routes of cold air from the spheric northern annular mode (NAM), also tends to be northern polar region to lower latitudes: namely, the in its negative phase when there are more frequent ‘‘East Asian stream’’ and the ‘‘North American strong cold air outbreaks in the midlatitudes of Eurasia stream.’’ Shoji et al. (2014) conducted a comprehensive and North America (Thompson and Wallace 1998, 2001; isentropic diagnosis of East Asian cold air outbreaks Wettstein and Mearns 2002; Cohen et al. 2010). Ac- within the cold air branch of the meridional mass cir- companying these leading modes is the oscillation of culation. Yu et al. (2015) constructed a mass circulation extratropical zonal mean zonal wind in the troposphere, index (denoted as WB60N) to measure the intensity of which was termed the ‘‘index cycle’’ to link variations of the warm air branch of the meridional mass circulation the westerly jet to cold air outbreaks in the pioneering and showed that changes of the index can be a precursor work of Namias (1950). for the cold air outbreaks in midlatitudes. They also Several promising precursors to winter cold air out- showed that there exist two dominant geographical breaks have also been found in the stratosphere. The patterns of temperature anomalies during the cold air works of Baldwin and Dunkerton (1999), Wallace discharge period (or 1–10 days after a stronger mass (2000), and Thompson et al. (2002) indicate that cold air circulation across 608N). One represents cold anomalies outbreaks tend to occur more frequently over the mid- mainly in the midlatitudes of both North America and latitudes in the period of 1–2 months after a weak Eurasia, and the other represents cold anomalies mainly stratospheric polar vortex event. Cai (2003) found that over only one of the two continents accompanied with cold surface temperature anomalies tend to take place abnormal warmth over the other continent. Yu et al. underneath the intrusion zone of high isentropic po- (2015) mainly focus on statistical evidence of the robust tential vorticity (IPV) into the troposphere from the relation between the strengthening of meridional mass stratosphere. Kolstad et al. (2010) found that cold tem- circulation across 608N and cold air outbreaks in the perature anomalies over the southeastern United States midlatitudes but do not examine the spatiotemporal tend to appear within 1–2 weeks after the peak dates of patterns of anomalous meridional mass circulation and weak vortex events, whereas the cold anomalies over Eurasia seem to appear at the inception of weak vortex events. In addition, continental-scale cold temperature 1 The poleward and equatorward branches of the meridional anomalies are correlated with the easterly phase of the mass circulation are defined in terms of the zonally integrated mass equatorial stratospheric quasi-biannual oscillation (e.g., fluxes. Since the zonally integrated mass fluxes are poleward in the Thompson et al. 2002; Cai 2003). upper isentropic layers and equatorward in lower isentropic layers, The pioneering work of Johnson (1989) and many they are also referred to as the warm and cold air branches in the subsequent studies [e.g., Cai and Shin (2014) and ref- literature (e.g., Cai and Shin 2014). Within each of the two branches, there exist both poleward warm airmass fluxes in some erences therein] established a hemisphere-wide single- longitudinal sectors and equatorward cold airmass fluxes in others. cell model for meridional mass circulation. The mass This is particularly true in the extratropics, where the meridional circulation is a thermally direct circulation that connects mass circulation is mainly driven by baroclinically amplifying waves.

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3216 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

TABLE 1. A list of the derived indices and variables.

Derived indices and variables Variables needed for calculation Reference

Isentropic mass u; us, Ps Pauluis et al. (2008) MF* y, u; ys, us, Ps Cai and Shin (2014) u_ u u_ u VF , ; s, s, Ps Yu et al. (2014) CMI y, u; ys, us, Ps Iwasaki et al. (2014) WB60N index MF at 608N Yu et al. (2015) Temperature area indices Anomalies and local standard deviations of surface Cai (2003), air temperature Yu et al. (2015) WAI Geopotential height Zhang et al. (2013), Yu et al. (2014)

* MF(l, f, Qn) is obtained by summing up meridional mass fluxes in all layers at a given grid box for which the potential temperature u is between Qn and Qn11. The meridional mass flux in a grid box at u, denoted as f (l, f, s; u), is evaluated on the sigma coordinate as the product of mass [DsPs(l, f)/g] and meridional wind [y(l, f, s; u)], where Ds 5 0.05 and Ps is the surface pressure [see Yu et al. (2015) for more details]. their linkages to the anomalous changes in synoptic Other variables and indices used in this study are circulation systems that are directly connected with cold calculated from variables available in daily ERA- air outbreaks. The primary objective of this study is to Interim data following the corresponding references examine the dominant spatiotemporal circulation pat- listed in Table 1. Each of the 32 winters lasts for 120 days terns of the meridional mass circulation and indicate from 1 November of the current year to 28 February of their roles in setting up the preferred routes for cold air the next year. The daily climatological-mean fields are outbreaks in the midlatitudes during anomalous mass obtained by averaging the data across the 32 years circulation events. (1979–2011) for each calendar day from 1 November to This paper is organized as follows. Section 2 describes 28 February. Daily anomalies are obtained by removing the dataset and indices used in this study and outlines the daily climatology from the total fields. Next, the the analysis procedures. In section 3, we examine the life physical meaning of each of the calculated variables and cycle of anomalous WB60N events and the related indices is briefly introduced. temporal evolution of surface temperature anomalies. We have used the same methods in Yu et al. (2015) Section 4 presents the temporal and spatial patterns of [also see Pauluis et al. (2008) and Cai and Shin (2014)]to various circulation anomalies associated with anoma- calculate variables associated with mass circulation (in- lous WB60N events, including the adiabatic and diabatic cluding air mass, mass tendency, and meridional and components of mass transport anomalies, the preferred vertical mass fluxes in isentropic layers) from daily routes of cold air outbreaks, the anomalies in surface fields. We have preselected 13 potential temperature synoptic systems, and tropospheric wave activities. surfaces Qn (n: 1–13): 260, 270, 280, 290, 300, 315, 330, Conclusions are provided in section 5. 350, 370, 400, 450, 550, and 650 K. Vertical mass fluxes are defined at all Qn that satisfy Qn .Qs, where Qs is the surface potential temperature varying with location and 2. Data and analysis procedures time (note that the mass flux across Qs is always zero, The data fields used in this study include the temper- since we only consider dry air mass in this study). All ature (Ts), pressure (Ps), and meridional wind (ys)atthe other variables are defined in the 12 layers between Qn surface, and the three-dimensional air temperature (T), and Qn11 surfaces plus two additional layers: one is the geopotential height (z), and meridional wind (y), which surface layer, which accounts for all mass between the are derived from the daily ERA-Interim data for the 32 ground and the minimum of Qn that satisfies Qn .Qs, winters from 1 November 1979 to 28 February 2011 and the other is the top layer, which accounts for all air (ECMWF 2012; Simmons et al. 2006; Dee et al. 2011). mass above 650 K. We use the bottom surface of each The data fields are on 1.58 latitude 3 1.58 longitude layer (Qn, Qs for the surface layer) in referencing the grids and on 37 pressure levels spanning from 1000 to variables defined in these isentropic layers. 1 hPa. Three-dimensional and surface potential tem- We denote a zonally integrated field using angle perature (u and us) fields are obtained from the tem- brackets. For example, the zonally integrated meridio- perature fields at pressure levels. Total diabatic heating nal mass flux (MF) is denoted as hMFi which is a func- u_ u_ f Q h i fields ( and s) are calculated from 6-hourly three- tion of , n, and t. Positive values of MF correspond dimensional and surface temperature, wind, and ver- to poleward mass transport on day t across the latitude f tical motion fields. within the two adjacent isentropic surfaces (Qn and

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3217

Qn11), and negative values of hMFi correspond to equa- heat content in Iwasaki et al. (2014), to measure the torward transport. On average, negative values of hMFi degree of coldness of the air mass within the cold air mainly appear in the lower troposphere within the equa- branch. To obtain the total cold mass intensity within the torward cold air branch of the meridional mass circula- cold branch, we first calculate the product of the air mass tion, while positive values mainly appear in upper layers in each isentropic layer in this grid box and the de- within the poleward warm air branch (see Fig. 1a in Cai parture of its potential temperature from 280 K and then and Shin 2014). For the zonally integrated vertical mass sum them up from the surface layer to the layer of 280 K, fluxes (VF), denoted as hVFi, positive values correspond which roughly corresponds to the layer that separates to upward mass fluxes across the isentropic surface Qn the warm and cold branches of the mass circulation in because of diabatic heating, and negative values corre- the extratropics (Iwasaki et al. 2014; Cai and Shin 2014). spond to downward mass fluxes because of diabatic The absolute value of the sum is defined as the CMI as a cooling. Positive values of hVFi are observed mainly in function of time. In general, the larger the CMI is, the the tropics and the surface layer in the extratropics; colder the air is in the cold air branch. whereas negative values are found outside of the tropics The level that separates the warm and cold branches above the surface layer (see Fig. 1b in Cai and Shin 2014). at latitude f and at each day t is identified by searching Combining the vertical and meridional mass flux fields, we for the isentropic level Qn*(f, t) such that the vertical obtain the isentropic meridional mass circulation as a sum of hMFi for all n , n*orQn ,Qn* reaches its function of latitude and isentropic surface level at each maximum negative value. Generally, Qn* decreases with day. The net mass change due to convergence of meridi- latitude, which reflects the latitudinal variation of onal and vertical mass fluxes is measured by the daily stratification in the troposphere. The value of tendency of the total mass between two adjacent isen- Qn*(f 5 608N, t) mainly lies between 270 and 290 K. tropic surfaces (Qn and Qn11) within each latitude band f. Following Yu et al. (2015), the WB60N index is con- Following Iwasaki et al. (2014), we adopt the cold structed to measure the intensity of the poleward warm mass intensity (CMI), which was termed as negative air branch crossing 608N at a given time according to

8 9 t t 650K < 650K = å h i f 5 8 Q 2 å h i f 5 o Q MF ( 60 N, n, t) MF ( 60 N, n, t) Q 5Q :Q 5Q ; n n* n n* WB60N(t) 5 . (1) SDW

The overbar with the superscript t denotes a 7-day high latitudes in the regressed surface temperature running mean centered at time t. The curly braces the anomalies against the mass circulation index. However, time average over the 32 winters from 1979 to 2011 for the choice of 608N yields the largest amplitude of tem- each calendar day between 1 November and 28 Feb- perature anomalies in such a seesaw pattern. ruary, which produces a 120-day time series of winter- Note that the net meridional mass transport in a given season daily climatology of total mass transport crossing layer crossing a given latitudinal circle in the extra- 608N into the polar region in the poleward warm tropics is always a residual between the equatorward 2 air branch. The quantity SDW 5 16.4 3 109 kg s 1,rep- mass transport of cold air behind the trough and the resents the standard deviationst of the time series of poleward mass transport of warm air in front of the 650K trough, because the extratropical portion of the merid- åQ 5Q hMFi(f 5 608N, Qn, t) . Based on the analysis n n* of the WB60N index in Yu et al. (2015), variations of ional circulation is driven by large-scale westward-tilted the anomalous mass transport into the polar region by baroclinic waves (Johnson 1989). To investigate the the warm air branch, as well as that out of the polar preferred pathways of the equatorward and poleward region by the cold air branch, can be represented by this airmass transport within the cold branch, we also cal- index because of the synchronization of the two culate the 7-day running-mean meridional mass trans- l f branches in terms of their timing and intensity. port at longitude and across latitude , denoted as The mass circulation index is deﬁned at 608N because MFCB, according to the cold polar air mainly resides north of 608N inside the t Q polar region (Iwasaki et al. 2014). We have also tested n*21 8 8 l f 5 å l f Q the choices of latitudes from 50 to 70 N and always MFCB( , , t) MF( , , n, t) . (2) Q 5Q found a similar seesaw pattern between mid- and n s

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3218 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

2 By definition, the zonal summation of MFCB(l, f, t) and the mean peak intensity of WB60N events is also is equal to the net equatorward mass transport cross- close to 1.5 standard deviations below the climatology. It ing latitude f within the cold branch. The anomaly should be noted that, though the composite temporal 1 2 fields of MFCB can be obtained by removing its cli- evolutions of WB60N and WB60N events are very matological annual cycle. Positive anomalies of MFCB similar to each other, individual events exhibit some no- 1 2 represent either a reduction in the equatorward cold ticeable asymmetries between WB60N and WB60N airmass transport or an increase in the poleward warm events. Specifically, the time span of the circulation 1 airmass transport crossing a specific latitude f through strengthening period before the peak times of WB60N a 1.58 longitude sector centered at l in the lower tro- events seems to be shorter than the weakening period posphere. The opposite is true for negative anomalies after, while the time span of the circulation weakening 2 of MFCB. period before the peak times of WB60N events is also 1 As in Yu et al. (2015), warm and cold temperature relatively longer; the duration of WB60N is generally 2 1 area indices are defined to measure the spatial extent or shorter than WB60N ; and the peak intensity of WB60N 2 the percentage of area occupied by warm or cold events tends to be stronger than that of WB60N events. anomalies that exceed the 0.5 local standard deviation. In the remainder of the paper, we report composite The cold area index for midlatitude (258–608N) is de- temporal evolutions of temperature area indices and noted as CM, and that for high latitude is denoted as CH. circulation anomalies according to the timeline in the The corresponding warm area indices are WH and WM, range from 27 days to 7 days relative to the positive and 1 respectively. negative peak dates of individual WB60N (157) and 2 WB60N (161) events. The choice of [27, 7] days is made because it is slightly longer than the average time 3. Composite anomalous WB60N events and the scale of individual anomalous WB60N events. The pe- associated surface air temperature anomalies riod from 27 to 7 days relative to the peak times of There are a total of 157 episodes of strong mass cir- anomalous WB60N events can cover the temporal span culation in the 32 winter seasons that have peak values of the majority of individual anomalous WB60N events of WB60N exceeding 0.5 or a poleward mass transport (black curves in Fig. 1). Below, we will plot these com- into the upper polar region exceeding half a standard posite fields sequentially from [27, 7] days relative to 1 deviation above its climatological value for more than the peak dates of WB60N events to [27, 7] days rela- 2 three days. For ease of reference, we refer to these as tive to the peak dates of WB60N events. But it should 1 1 WB60N events. There are a total of 161 episodes of be noted that (i) the duration of individual WB60N or 2 weak mass circulation that have minimum values of WB60N events varies as indicated in Fig. 1; (ii) dis- 1 WB60N below 0.5 or a poleward mass transport into playing the composite WB60N events sequentially to- 2 the upper polar region below the climatological value gether with the composite WB60N events does not by half a standard deviation, and these are referred to necessarily mean that the actual evolution from the 2 1 2 as WB60N events. The black curves in Fig. 1 depict WB60N to the WB60N takes place in around 30 days; 1 the temporal evolutions of these individual WB60N and (iii) the composite evolution may include some 2 1 2 and WB60N events, and the red curves are the aver- overlapping between the WB60N and the WB60N 1 2 ages of these WB60N (Fig. 1a)andWB60N (Fig. 1b) events when the actual transition between them is very 1 events, referred to as the composite WB60N and abrupt and their duration is shorter than 15 days. 2 WB60N events, respectively. It is found that the du- Displayed in Figs. 2a and 2b are the temporal evolu- 1 2 ration of both WB60N and WB60N events has a tions of composite-mean daily tendencies of the tem- relatively wide range, from 3 days up to 2–3 weeks. On perature area indices CH (solid blue), CM (dashed blue), 1 average, it takes about 5 days for a WB60N event to WH (solid red), and WM (dashed red) in the period 1 reach its peak value from the climatological-mean cir- from 27 to 7 days, relative to the peak dates of WB60N 2 culation intensity (i.e., the value of 0) and another 5 days (Fig. 2a) and WB60N (Fig. 2b) events. The tendencies to return to 0 from the peak intensity. The same can be of WH and CM evolve highly in phase with WB60N, 2 said for a WB60N event. Therefore, the average time exhibiting their positive maximums at the peak dates of 1 2 1 scale of both WB60N and WB60N events is about WB60N and their negative maximums at the peak 1 2 10 days. The mean peak intensity of WB60N events is dates of WB60N , while the tendencies of CH and WM close to 1.5 standard deviations (one standard deviation are generally out-of-phase with WB60N. The positive corresponds to poleward mass transport into the polar peak of the tendency of CM and the negative peak of the atmosphere in the warm air branch at a rate of 16.4 3 tendency of WM correspond to the timing of the onset of 2 2 109 kg s 1) above the climatology (about 79.8 3 109 kg s 1), cold air outbreaks in the midlatitudes, which themselves

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3219

1 2 FIG. 1. Temporal evolutions of (a) 157 WB60N and (b) 161 WB60N events centered at their peak times (black curves). Note that these black curves only cover the temporal spans 1 during which the WB60N index is continuously above zero (for WB60N )orbelowzero 2 (for WB60N ). The red curves correspond to their composite means. The ordinate represents the value of the WB60N index, and the abscissa is the lag days from the time of peak amplitude.

1 coincide with the peak dates of WB60N events. By the meridional mass circulation results in the maximum of 1 same token, the negative peak of the tendency of CM CM and minimum of WM at ;5 days after the WB60N 2 and the positive peak of the tendency of WM correspond peak dates, and vice versa after the WB60N peak to the timing of the demise of the cold period in the dates (Figs. 2c,d). The dates with maximums of the 2 midlatitudes, coinciding with the peak dates of WB60N percentage area indices for CM and WH and minimums events. Physically speaking, stronger meridional mass of CH and WM are referred to as the mature dates of circulation is associated with more cold airmass trans- cold air outbreaks. The mature dates of the percentage port from the polar region to the midlatitudes, which is area indices for the midlatitudes (CM and WM) appear responsible for an increase in the area occupied by cold to be 1–2 days earlier than those (CH and WH)forthe surface air temperature anomalies in the midlatitudes high latitudes. and a decrease in the area occupied by warm surface air Shown in Figs. 3a–f are composite maps of surface air 1 temperature anomalies. The reverse is true when the temperature anomalies in different phases of WB60N meridional circulation is weaker. The continuous events (maps along the right half of the circle around cooling tendency during the strengthening of the Fig. 3g, which shows the stationary wave component of

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3220 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

21 1 FIG. 2. Composites of temporal tendencies of temperature area indices (% day ) during (a) WB60N events and 2 1 2 (b) WB60N events, and temperature indices (%) during (c) WB60N events and (d) WB60N events in the period

from 27to17 days from the peak dates of anomalous WB60N events. Solid (dashed) blue curves are for CH (CM); and solid (dashed) red curves are for WH (WM). Dots denote composites that are statistically signiﬁcant above the 95% conﬁdence level.

1 the climatological-winter-mean surface air temperature polar region. At the peak time of WB60N , when the 2 ﬁeld) and WB60N events (maps along the left half). equatorward mass transport across the polar circle is These composite maps compliment the results shown in strongest, cold surface temperature anomalies begin to Fig. 2 with information on both the temporal evolution move southward, while warm anomalies begin to appear and the geographical pattern of surface air temperature over the Nordic seas and Bering Sea (Fig. 3b). During 1 anomalies. The quadrature relation of surface air tem- the week after the peak dates of WB60N events, cold perature anomalies with the WB60N index revealed in surface temperature anomalies intensify and gradually Fig. 2 can be seen vividly in the temporal evolution of spread over the entire midlatitudes, with the coldest the geographical pattern of surface air temperature centers over the two continents. Meanwhile, warm anomalies during different phases of anomalous temperature anomalies also intensify and spread over WB60N events. The reversal of the opposite sign of the polar region, although the warmest centers are still surface air temperature anomalies between the polar over the Nordic seas and Bering Sea. Conversely, during 2 region and the midlatitudes takes place around the peak the week before the peak dates of WB60N events, 1 dates of WB60N events (Fig. 3a vs Fig. 3c) and warm temperature anomalies over the Nordic seas and 2 WB60N events (Fig. 3d vs Fig. 3f). Prior to the peak Bering Sea, as well as cold temperature anomalies over 1 dates of WB60N events, warm anomalies are still the two continents, begin to gradually diminish. On the 2 prevalent over the midlatitudes of Eurasia and North peak dates of WB60N events, the midlatitudes of East America, while cold anomalies are present over the Asia and the northwest Paciﬁc Ocean are occupied by

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3221

FIG. 3. Composite surface air temperature anomalies (K) in various phases of anomalous WB60N events in the domain of 258–908N 1 clockwise from top right: (a) 27to22, (b) 21 to 1, and (c) 2 to 7 days from the peak dates of WB60N events; and (d)–(f) as in (a)–(c), but 2 for WB60N events. Slashed areas denote the composites are statistically significant above the 95% confidence level. (g) The stationary wave component of the 32-yr-, November–February (NDJF)-mean surface air temperature field. The contours for (a)–(f) are shown in the color bar in the bottom left, and those for (g) are in the bottom right. warm temperature anomalies, while part of the polar 4. Circulation anomalies associated with region north of Eurasia becomes anomalously cold anomalous WB60N events (Fig. 3e). During the week after the peak dates of 2 WB60N events, the entire midlatitudes are dominated In this section, the temporal evolutions of the spatial by warm temperature anomalies, with the warm centers patterns of circulation anomalies in various fields asso- over the two continents, whereas cold surface temper- ciated with anomalous WB60N events are examined. ature anomalies return to the polar region. The week The identification of the spatial patterns of circulation 2 after the peak dates of WB60N events can be referred anomalies that evolve with the anomalous WB60N to as the ‘‘cold air charge period,’’ and the week after the events helps to provide a more physics-based explana- 1 peak dates of WB60N events corresponds to the ‘‘cold tion as to why the onset dates of cold air outbreaks in the 1 air discharge period,’’ as in Yu et al. (2015). midlatitudes coincides with the peak dates of WB60N

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3222 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72 events, and the mature phase of cold air outbreaks tends the midlatitudes to the polar region in the upper layer 1 to appear a week after the peak dates of WB60N events. (above 280 K) and more cold air from the polar region to the midlatitudes below. After the peak dates, anom- a. Meridional mass circulation anomalies alies of both the meridional mass flux and the mass Figure 4 shows the composite anomalies of the zonally tendency begin to weaken (Fig. 4d) and their signs be- integrated mass fluxes (vectors), the mass tendency come reversed at 5–7 days after the peak dates of 1 (contours), and the zonally integrated mass itself (shad- WB60N events (Fig. 4e). The mass anomaly pattern, 1 2 ing) during the WB60N and WB60N events. Note that however, remains unchanged throughout the week after the total meridional mass fluxes always tend to be pole- the peak dates because of the temporal delay of the mass ward (positive values in the Northern Hemisphere) in the field with respect to the mass transport. Moreover, the upper atmosphere within the warm air branch and equa- meridional extent and the intensity of the mass anom- torward (negative values in the Northern Hemisphere) alies reach their maximums about 4–5 days after the near the surface within the cold air branch. Therefore, peak dates. positive–negative meridional mass flux anomalies in Opposite patterns of meridional mass circulation, upper levels imply stronger–weaker poleward mass mass tendency, and mass anomalies are found during the transport, whereas, in lower levels, negative–positive negative half of anomalous WB60N events (Figs. 4f–j). mass flux anomalies are indicative of stronger–weaker Comparing Figs. 4a–e with Figs. 3a–e and Figs. 4f–j with equatorward mass transport. Figs. 3f–j, it is evident that mass anomalies induced by It is apparent that both the warm and cold branches of equatorward mass transport anomalies in the lower- the meridional mass circulation begin to intensify a few tropospheric layers are closely linked to surface tem- 1 days before the peak time of WB60N events, as in- perature anomalies in the extratropics, in terms of both dicated by the gradual decrease in negative meridional timing and intensity. mass flux anomalies in upper levels and positive mass It should be noted that, following the strengthening or flux anomalies in lower levels in the extratropics the weakening of the meridional mass transport in the (Figs. 4a,b). During this period, the mass tendency extratropics, the vertical mass flux anomalies can also anomalies above 280 K change from negative to positive contribute to mass tendency and mass anomalies. Dur- 1 in the high latitudes and from positive to negative in ing WB60N events, the strengthened meridional mass the midlatitudes, but the opposite is found below 280 K. transport is followed by anomalous upward mass trans- This is a result of the simultaneous strengthening of the port south of 608N and anomalous downward mass poleward mass transport in the warm air branch and transport in the north within the week after the peak 1 equatorward mass transport in the cold air branch. Since time of WB60N events (see the vertical component of changes of mass anomalies must take place after mass flux anomalies in Figs. 4c–e). This indicates that changes in mass transport that cause mass tendency in more meridional mass transport of the warm air into the the first place, mass anomalies north of 658N are still upper polar region is accompanied by stronger radiative negative above 280 K and positive below 280 K, and cooling there, while more transport of the cold polar air those south of 658N are still positive above 280 K and mass into the lower midlatitudes is associated with negative below during the period of 2–7 days before the stronger diabatic heating from the ground. The opposite 1 peak dates of WB60N events. The strengthening of pattern of diabatic mass flux anomalies is found in the 2 the meridional mass circulation leads to a reversal of the WB60N events. Although these anomalous cross- pattern of the opposite signs of mass anomalies in the isentropic mass transports tend to cancel out part of meridional and vertical directions shown on the peak the mass and its tendency anomalies induced by the 1 dates (Fig. 4c). On the peak dates of WB60N events, anomalous meridional mass transport, the polarity of the poleward mass flux anomalies in the upper layer are the mass and its tendency anomalies is still determined the strongest, extending to the North Pole, while the by the polarity of the anomalous meridional mass equatorward flux anomalies in the lower layer are also transport. the strongest, extending southward up to 308N. The b. Anomalies of the CMI strongest meridional mass circulation on the peak dates causes the largest positive (negative) mass anomalies To confirm the close relation between the mass above 280 K, and the largest negative (positive) anom- anomalies in lower isentropic layers and surface tem- alies below, in the high latitudes (midlatitudes) within a perature anomalies, we next examine (Fig. 5) the tem- 1 few days after the peak dates of WB60N events poral evolution of the meridional profiles of anomalies (Fig. 4d). This indicates that the strengthened meridio- of the vertically integrated mass in layers below 280 K nal mass circulation has transported more warm air from and CMI (shading) and their daily tendencies (contours)

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3223

FIG. 4. Composite anomalies of zonally integrated mass fluxes (vectors), mass (shading), and mass tendency (contours) in different phases of anomalous WB60N events as a function 1 of latitude (abscissa) and isentropic level (ordinate): (a)–(e) WB60N events and 2 2 (f)–(j) WB60N events. The length of the reference vector corresponds to 10 3 109 kg s 1; the contours for the shading are in 1014 kg; and the mass tendency contour interval is 2 0.2 (109 kg s 1), with solid contours for positive values and dashed contours for negative values. Composites not exceeding the 95% confidence level in terms of statistical significance are omitted for the mass and mass flux anomalies, and displayed as thin contours for the mass tendency anomalies.

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3224 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

14 FIG. 5. Composite anomalies of the zonally integrated mass under 280 K (shading; 10 kg) and its tendency (contour interval: 0.5 3 2 1 109 Kkgs 1) as a function of latitude (abscissa) and number of days from the peak time (ordinate) of (a) WB60N events and 2 (b) WB60N events. (c),(d) As in (a),(b), but for the anomalies of the CMI (shading; 1014 K kg) and its tendency (contour interval: 10 3 2 109 Kkgs 1). Composites not exceeding the 95% confidence level in terms of statistical significance are omitted for the anomalies represented by shading, and are plotted in thin contours for those indicated by contours. in the period from 27 to 7 days, relative to the peak polar region, implying anomalously cold conditions in 1 2 dates of WB60N and WB60N events. It can be seen the midlatitudes and warm conditions in the high lat- that the temporal evolution and the latitudinal profile of itudes.TheoppositesituationcanbeseeninFigs. 5c 2 the mass and mass tendency anomalies (Figs. 5a,b) are and 5d for WB60N events: the weakening of the highly similar to their counterparts of the CMI and CMI meridional circulation leads to a negative cold airmass tendency anomalies (Figs. 5c,d). The meridional dipole and CMI anomalies in the midlatitudes, as well as a pattern of the cold airmass depletion tendency in the positive cold airmass and CMI anomalies in the high high latitudes and the cold airmass accumulation ten- latitudes. dency in the midlatitudes reaches its peak amplitude as a 1 c. Anomalies in surface pressure systems and routes WB60N event approaches the peak time, as does the of the meridional mass transport near the surface meridional dipole pattern of the negative CMI tendency in the high latitudes and the positive CMI tendency in Figure 6 shows the meridional mass transport within the midlatitudes. This clearly indicates that the increase the cold air across 608N [i.e., MFCB(l, f 5 608N, t), de- of air mass in lower isentropic layers indeed corresponds fined in Eq. (2)]. It is seen that, under winter-mean to the increase of CMI, and vice versa. After the peak of conditions (MFCB; Fig. 6a), the two main routes for the 1 WB60N events, positive anomalies of both cold air Arctic cold surface air to enter the midlatitudes are via mass and CMI appear in the midlatitudes, and negative East Asia and eastern North America, and the two main anomalies of cold air mass and CMI appear over the routes for warm surface air to enter the Arctic lie over

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3225

FIG. 6. The longitudinal profile of the total meridional mass transport within the cold air 2 branch at 608N (109 kg s 1): (a) climatological winter (NDJF) mean and (b),(c) composite 1 anomalies in the period from 27to17 days (ordinate) from the peak time of WB60N and 2 WB60N events, respectively. Slashed areas denote the composites that are statistically significant above the 95% confidence level. the two oceans. A secondary route for the Arctic cold air stronger than the poleward transport, the zonally in- lies in western Europe, which is paired with a secondary tegrated net mass transport within the cold air branch is route over central Eurasia for the poleward transport of always equatorward. Next, let us examine the composite 1 warm air. Because the equatorward mass transport is anomalies of MFCB in various phases of the WB60N

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3226 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

FIG.7.AsinFig. 3, but for the wave ﬁelds of geopotential height at 1000 hPa (m).

2 (Fig. 6b) and WB60N (Fig. 6c) events. It is seen that the weakening of the equatorward mass transport over East simultaneous strengthening of the equatorward mass Asia and eastern North America and the poleward transport over East Asia and North America, and the transport over the Pacific and Atlantic Oceans, as well as poleward transport over the Pacific and Atlantic the strengthening or resumption of the secondary routes Oceans, takes place in the period from a few days before of Arctic cold air via western Europe and warm air via 1 the peak time of WB60N events to several days after. central Eurasia. The mass transport via the secondary winter-mean Next, we use the winter climatological-mean maps of routes seems to be greatly weakened or even reversed the stationary wave fields of geopotential height at 1 during the week after the peak time of WB60N events. 1000 hPa (z1000; Fig. 7g) and meridional mass fluxes Specifically, the anomalous equatorward mass transport within the cold air branch (Fig. 8g) to reference the of Arctic cold air is via central Eurasia, and the anom- winter permanent surface pressure systems and clima- alous poleward mass transport is via western Europe tological mean routes of winter cold outbreaks. The 1 during the week after the peak time of WB60N events. extratropical portion of the meridional mass circulation The reverse pattern is found within the week after the is driven by westward-tilted baroclinic Rossby waves 2 peak time of WB60N events: i.e., the simultaneous (e.g., Johnson 1989). Accordingly, it is expected based

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3227

9 21 FIG.8.AsinFig. 3, but for (a)–(f) the anomalies of the total meridional mass transport within the cold air branch (10 kg s ) and (g) the corresponding winter climatological mean. on the geostrophic balance that, in the Northern favors the two main routes via the eastern half of the Hemisphere, the equatorward portion of the winter Pacific Ocean, the Atlantic Ocean, and a secondary climatological mean MFCB field will lie on the east side route via central Eurasia. The main routes are well of the winter-mean surface high pressure centers or on coupled with the four permanent surface systems— the west side of the winter-mean surface low pressure namely, the Siberian high, the North American high, centers, and vice versa for the poleward portion of the the Aleutian low, and the Icelandic low—whereas the winter-mean MFCB field (Fig. 7g vs Fig. 8g). As in pair of the secondary routes is associated with the Si- Fig. 6a, which is evaluated only at 608N, Fig. 8g confirms berian high and the Azores high. that, on average, the polar cold air prefers to advance Next, we discuss the composite wave fields of geo- 0 equatorward via the entire extratropical latitude span of potential height anomalies at 1000 hPa (z1000) and the East Asia and North America, with a secondary route anomalies of meridional mass fluxes within the cold air 0 via western Europe. Meanwhile, the poleward mass branch (MFCB) obtained by averaging from one week 1 transport from the midlatitudes to the polar region before to one week after the peak dates of all WB60N

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3228 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

2 events (Figs. 7a–c and 8a–c) and WB60N events climatological cold air route through western Europe, (Figs. 7d–f and 8d–f). The overall in-phase relationship which is responsible for normal temperatures there 0 between z1000 shown in Figs. 8b–c and z1000 in Fig. 8g is during this period. The evidence provided above helps indicative of the strengthening of the winter-mean sur- to explain physically why cold anomalies mainly appear 1 face pressure system on the peak dates of WB60N over the midlatitude regions of Eurasian and North events and during the week after. Specifically, positive American continents, while warm anomalies in the polar 0 anomalies of z1000 are found over Eurasia with maxi- region exhibit maximums over the northern sides of the mums over northwestern portions of central Eurasia, Pacific and Atlantic oceans when the meridional mass which superpose the northwestern part of the winter- circulation is anomalously strong, as conjectured in Yu mean Siberian high. It follows that the Siberian high, et al. (2015). which has been recognized as the key weather system According to Figs. 6b and 8a–c, the timing of the peak initiating cold air surges over East Asia (e.g., Ding 1990; intensity of the anomalous equatorward mass fluxes Zhang et al. 1997), is anomalously strong during peak through the three cold air routes and that of the anoma- 1 dates of WB60N events. As reported in Walsh et al. lous poleward mass fluxes within the two warm routes are (2001), positive sea level pressure anomalies over the not exactly synchronized with one another. The equa- 0 Alaska–Yukon border are closely linked to cold air torward MFCB over North America and that over East outbreaks in North America. Again, we see positive Asia reach their peak intensities during the peak dates of 0 1 anomalies of z1000 over the Alaska–Yukon border WB60N events and then weaken significantly after- 1 0 around the peak time of WB60N events, though the ward. The equatorward MFCB via the route of central amplitude is weaker than that over the Siberian high Eurasia reaches its peak intensity a few days later, after 1 region. Meanwhile, the Aleutian low and the Icelandic the peak dates of WB60N events. The intensity of the 0 low are anomalously deepened, as manifested by the poleward MFCB through two oceans also peaks a few 0 negative z1000 there. Therefore, the strengthening of the days after the peak dates. Such a delay of the equator- 0 permanent surface pressure systems in winter, which ward MFCB through central Eurasia, with respect to those determine the climatological-mean routes of winter cold through the other two main routes, explains why the outbreaks, acts to bring more polar cold air equatorward strengthening of the cold anomalies over central Eurasia on the east (high) side of high (low) pressure centers via slightly lags that over East Asia and North America. In 0 the two main continental routes into the midlatitudes addition, the poleward MFCB via the Atlantic and Pacific 0 and to carry more warm air poleward on the west (east) Oceans slightly lags the equatorward MFCB via the two side of high (low) pressure centers via oceanic routes continents. This explains why the polar region continues into the polar region. to get warmer after the peak intensity of cold anomalies Besides the strengthening of the winter surface pres- in the midlatitudes (see Figs. 2, 3). 2 sure systems, the center of the winter high pressure During WB60N events, the four winter permanent system over Eurasia appears to shift westward during surface pressure centers are weakened substantially, as 1 WB60N events. According to the composite-mean indicated by the overall out-of-phase relationship be- 0 0 MFCB shown in Figs. 8b and 8c and the longitudinal tween z1000 and z1000 (Figs. 7e,f vs Fig. 7g). The weak- 0 8 profile of MFCB at 60 N shown in Fig. 6b, such a shift ened surface pressure systems correspond to a weakens the secondary winter-mean poleward route for weakened equatorward mass transport on the east side warm air via central Eurasia, and even opens a new route of the high pressure centers and the west side of the low for cold polar air to enter central Eurasia (Figs. 8b,c pressure centers, responsible for anomalous warm con- vs Fig. 8g and Fig. 6b vs Fig. 6a). This accounts for the ditions over the eastern parts of Asia and the United 2 coldest composite-mean surface temperature anomalies States during WB60N events, as shown in Figs. 8e and 1 over central Eurasia in the period within one week after 8f and Fig. 6c. As the opposite case of WB60N , the 1 the peak dates of WB60N events (Figs. 3b,c). Previous northwestern portion of the winter high pressure system studies have also indicated that central Eurasia is a over central Europe weakens substantially (Figs. 7e,f vs critical region for trigging cold air surges over East Asia Fig. 7g). This causes the anomalous route for cold polar 1 through eastward-propagating wave trains (Wu and air to enter central Eurasia during WB60N events to Chan 1995; Ryoo et al. 2005; Hayasaki et al. 2006; become the route for warm air to enter the polar region Iwasaki and Mochizuki 2012; Iwasaki et al. 2014). The as the climatological mean condition (Figs. 8e,f vs Fig. 8g strengthening of the Icelandic low and Azores high, and Fig. 6c vs Fig. 6a), which is responsible for the which begins during the week before the peak time of maximum warmth in eastern Europe during the week 1 2 WB60N events and is indicative of the positive phase after the peak dates of WB60N events (Figs. 3e,f). The of the NAO, acts to weaken or possibly close off the weakening of the Icelandic low and Azores high,

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3229 corresponding to the negative phase of the NAO, begins between the WAI anomalies in the troposphere and the 1 as early as the end of WB60N events, or a week before WB60N index is probably as a result of the dependence 2 the peak time of WB60N events. This helps to of the net meridional mass transport not only on the strengthen or reopen the secondary winter-mean cold amplitude, but also the westward-tilting angle of the air route via western Europe and its companion sec- waves. Further investigation is needed to prove this ondary climatological-mean poleward route for warm conjecture. The temporal evolution of WAI anomalies 2 air via central Eurasia, as shown in Figs. 6a and 8g. This during WB60N events is similar, but with opposite sign. explains why temperature anomalies over western Eu- This confirms that it is the day-to-day variation of wave rope are farther below normal during the week before activities that drives the day-to-day intensity variation of 2 the peak dates of WB60N events, while cold tempera- the meridional mass circulation in the extratropics. ture anomalies over the rest of the two major continents gradually diminish. The weakening of the Icelandic low 5. Conclusions and Azores high is strongest around the peak dates of 2 WB60N events and continues during the week after. To identify the key physical and synoptic linkages However, the centers of the weakening gradually shift between the intensity variability of the meridional mass westward, away from western Europe (Figs. 7e,f vs circulation and surface cold air outbreaks in winter, we Fig. 7d), which prevents the latitudinal span of the cold air use daily ERA-Interim data (1979–2011) and investigate route from farther extending to the southern latitudes of the spatiotemporal variations of various circulation western Europe. As a result, only northern Europe still anomalies associated with the meridional mass circula- suffers from below-normal temperatures in this period tion intensity variability in winter (November–February). 0 (Figs. 3e,f). The corresponding patterns of z1000 (Figs. 7d–f) Following Yu et al. (2015), a daily mass circulation index, presents a westward shift of the center of the negative defined as the normalized zonally integrated poleward NAO mode, resembling the typical patterns identified in mass transport in upper isentropic layers across 608N(de- Walsh et al. (2001) for cold air outbreaks over western noted as WB60N), is used to represent the intensity of the and northern Europe. meridional mass circulation as a function of time. The four temperature area indices developed in Yu et al. (2015) d. Anomalies in planetary wave activities (CM, WM, CH,andWH) are used to measure the intensity In this section, the temporal evolution of wave activ- and polarity of temperature anomalies in the midlatitudes ities during the anomalous WB60N events is examined. (258–608N)andhighlatitudes(northof608N). It is found

Following Zhang et al. (2013) and Yu et al. (2014),we that the peak time of the positive tendency of CM and the define a wave amplitude index (WAI) as a function of negative tendency of WM coincide with the peak time of 1 latitude, pressure level, and time to represent the me- WB60N events, whereas the peak time of the negative ridional and vertical structure of wave amplitude at each tendency of CM and the positive tendency of WM coincide 2 time. At a given latitude, the WAI index is the standard with the peak time of WB60N events. Moreover, maxi- deviation of the departure of (total) geopotential height mum values of the CM index or minimum values of the WM from its zonal-mean value over the entire longitudinal index tend to occur in the period within one week after the 1 span of the latitude. Shown in Fig. 9 are the temporal peak dates of WB60N events, and vice versa within one 2 evolutions of the height–latitude structure of the WAI week after the peak dates of WB60N events. Otherwise 1 anomalies during the WB60N (Figs. 9a–e) and similar but opposite relationships are found for the CH and 2 WB60N (Figs. 9f–j) events. The WAI anomalies in WH indices during anomalous WB60N events. The results Fig. 9 have been normalized by their temporal standard further confirm the finding of Yu et al. (2015) that cold air deviation at each isobaric level. Thus, positive WAI outbreaks in the midlatitudes, as well as the out-of-phase anomalies represent higher zonal asymmetry of the geo- relation between surface temperature anomalies over the potential height field or larger wave amplitude, and vice high and midlatitudes, are closely linked to the day-to-day versa. Positive WAI anomalies in the extratropics first intensity variability of the meridional mass circulation. appear near the surface and in the low latitudes and then The strengthening of the meridional mass circulation is reach their maximum a few days after the peak time of accompanied by the intensification of atmospheric wave 1 WB60N with a poleward- and upward-propagation activities in the extratropics throughout the troposphere signal (Fig. 9c). This is well coupled with the strongest and the stratosphere. The wave activities become the 1 cold mass branch circulation in the troposphere in this strongest a few days after the peak time of WB60N period (Figs. 4c and 8b). Also, positive WAI anomalies events and exhibit a poleward and upward propagation. gradually emerge in the stratosphere from the peak time Intensification of wave activities is synchronized with a to one week after. Note that the slightly lagged relation strengthening of the poleward mass transport in the

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3230 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

1 FIG. 9. Composites of the normalized WAI anomaly in various phases of anomalous WB60N events 1 as a function of latitude (abscissa) and isobaric surface level (ordinate): (a)–(e) WB60N events; 2 (f)–(j) WB60N events. The normalization factor for the normalizedWAIisthesameforalllatitudesat the same isobaric level, equaling the domain mean of the standard deviations of WAI at each isobaric level. Slashed areas denote composites that are statistically signiﬁcant above the conﬁdence level. upper troposphere within the warm air branch and the overwhelms that transported downward across isentro- equatorward mass transport below within the cold air pic surfaces because of diabatic cooling, leading to a net branch. In upper layers, the air mass transported increase of warm air mass in the high latitudes. In the adiabatically from midlatitudes to the polar region lower layers, the air mass transported out of the polar

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC AUGUST 2015 Y U E T A L . 3231

2 region by the strengthened equatorward cold air branch before the peak dates of WB60N events, while negative dominates over that transported in from the upper temperature anomalies over the rest of the two major layers because of diabatic cooling, resulting in a net continents gradually diminish. The westward shift of the decrease of cold air mass in the high latitudes. The op- center of the negative NAO mode during and within the 2 posite situation is found in the midlatitudes. The week after the peak dates of WB60N events limits strengthened equatorward cold air branch results in a the latitudinal span of the cold air route via western Eu- net increase of cold air mass in the lower troposphere. rope. Consequently, only northern Europe still suffers Thus, the strengthened equatorward cold air branch is from below-normal temperature in this period. responsible for an anomalous increase of CMI and de- We wish to add that the results presented here do not crease of surface temperature in the midlatitudes but an change when we use an index that directly measures the anomalous decrease of CMI and increase of surface air cold air branch intensity, because the cold air branch is temperature in the high latitudes. Consequently, within nearly perfectly synchronized with the warm branch aloft in 1 one week after the peak time of WB60N events, there terms of both timing and intensity. The simultaneous are below-normal temperatures in the midlatitudes and strengthening of both warm and cold air branches is ac- above-normal temperatures in the high latitudes. companied with the strengthening of westward-tilted large- The (composite) mean surface circulation anomaly pat- scale baroclinic waves. Therefore, the stronger warm air tern associated with the strengthening of the meridional branch of the mass circulation is a robust indicator of a si- mass circulation generally resembles that of the winter- multaneous strengthening, in the cold air branch, of both the mean pattern, indicating a strengthening of the Siberian poleward mass transport of the midlatitudes warm air into and North American highs and their westward shifts to- the polar region along the climatological routes over oceans ward central Eurasia and the Alaska–Yukon region, as well and the equatorward mass transport of Arctic cold air along as the deepening of the Aleutian and Icelandic lows. The the climatological routes through lands, as shown in Figs. 6 strengthening of the permanent surface pressure systems, and 8. This is the basis for the existence of the physical causal which determine the climatological mean routes of winter relationship between the mass circulation in the warm air cold outbreaks, acts to bring more polar cold air mass branch and cold air outbreaks. Furthermore, the perspective equatorward on the east (west) side of high (low) pressure from the warm air branch aloft would allow us to extend the centers via continental routes into the midlatitudes, and to causal information for cold air outbreaks to a longer lead carry more warm air poleward on the west (east) side of time in our future studies, based on the poleward propaga- high (low) pressure centers via oceanic routes into the polar tion signal of temperature anomalies within the warm air region. The strengthening of the equatorward transport of branch, as elicited in Cai and Ren (2007) and Ren and Cai polar cold air via these climatological routes is responsible (2007), which provides a direct linkage of the variability for massive cold air outbreaks over East Asia and North of the WB60N index to tropical forcing anomalies. 1 America during the week after the peak time of WB60N 2 events. The reverse is found during WB60N events. Acknowledgments. YYY and RRC are supported by a In addition, the center of the winter high pressure sys- 1 research grant from the National Science Foundation of tem over Eurasia tends to shift westward during WB60N China (41430533, 91437105). MC is supported by grants events. Such a shift results in an opening of an anomalous from the National Science Foundation (AGS-1262173 route for cold polar air to enter central Eurasia, which is and AGS-1354834), the NOAA CPO/CPPA program responsible for the coldest composite mean surface tem- (NA10OAR4310168), and the DOE Ofﬁce of Science perature anomalies there in the period within one week 1 Regional and Global Climate Modeling (RGCM) pro- after peak dates of WB60N events. The strengthening of gram (DE-SC0004974). the Icelandic low and Azores high acts to close off the climatological-mean cold air route through western Eu- rope, which is responsible for the normal temperatures REFERENCES 1 there. Following the end of WB60N events, or within a 2 Baldwin, M. P., and T. J. Dunkerton, 1999: Propagation of the Arctic week before the peak dates of WB60N events, weak- Oscillation from the stratosphere to the troposphere. J. Geophys. ening of the Icelandic low and Azores high, which corre- Res., 104, 30 937–30 946, doi:10.1029/1999JD900445. sponds to the negative phase of the NAO, helps to Cai, M., 2003: Potential vorticity intrusion index and climate vari- strengthen or reopen the secondary winter-mean cold air ability of surface temperature. Geophys. Res. Lett., 30, 1119, route via western Europe. Accompanied is the strength- doi:10.1029/2002GL015926. ——, and R.-C. Ren, 2007: Meridional and downward propagation ening of the secondary winter-mean warm air route via of atmospheric circulation anomalies. Part I: Northern central Eurasia. This explains why temperatures over Hemisphere cold season variability. J. Atmos. Sci., 64, 1880– western Europe are farther below normal during the week 19, doi:10.1175/JAS3922.1.

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC 3232 JOURNAL OF THE ATMOSPHERIC SCIENCES VOLUME 72

——, and C.-S. Shin, 2014: A total flow perspective of atmospheric Ren, R.-C., and M. Cai, 2007: Meridional and vertical out-of-phase mass and angular momentum circulations: Boreal winter mean relationships of temperature anomalies associated with the state. J. Atmos. Sci., 71, 2244–2263, doi:10.1175/JAS-D-13-0175.1. Northern Annular Mode variability. Geophys. Res. Lett., 34, Cellitti, M. P., J. E. Walsh, R. M. Rauber, and D. H. Portis, 2006: L07704, doi:10.1029/2006GL028729. Extreme cold air outbreaks over the United States, the polar Rogers, J. C., and H. van Loon, 1979: The seesaw in winter tem- vortex, and the large-scale circulation. J. Geophys. Res., 111, peratures between Greenland and northern Europe. Part II: D02114, doi:10.1029/2005JD006273. Some oceanic and atmospheric effects in middle and high Cohen, J., J. Foster, M. Barlow, K. Saito, and J. Jones, 2010: Winter latitudes. Mon. Wea. Rev., 107, 509–519, doi:10.1175/ 2009–2010: A case study of an extreme Arctic Oscillation event. 1520-0493(1979)107,0509:TSIWTB.2.0.CO;2. Geophys. Res. Lett., 37, L17707, doi:10.1029/2010GL044256. Ryoo, S.-B., W.-T. Kwon, and J.-G. Jhun, 2005: Surface and upper-level Colucci, S. J., and J. C. Davenport, 1987: Rapid surface anticyclo- features associated with wintertime cold surge outbreaks in South genesis: Synoptic climatology and attendant large-scale cir- Korea. Adv. Atmos. Sci., 22, 509–524, doi:10.1007/BF02918484. culation changes. Mon. Wea. Rev., 115, 822–836, doi:10.1175/ Shoji, T., Y. Kanno, T. Iwasaki, and K. Takaya, 2014: An isentropic 1520-0493(1987)115,0822:RSASCA.2.0.CO;2. analysis of the temporal evolution of East Asian cold air out- Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: breaks. J. Climate, 27, 9337–9348, doi:10.1175/JCLI-D-14-00307.1. Configuration and performance of the data assimilation system. Simmons, A., S. Uppala, D. Dee, and S. Kobayashi, 2006: ERA-Interim: Quart. J. Roy. Meteor. Soc., 137,553–597,doi:10.1002/qj.828. New ECMWF reanalysis products from 1989 onwards. ECMWF Ding, Y., 1990: Build-up, air mass transformation and propagation Newsletter, No. 110, ECMWF, Reading, United Kingdom, 26–35. of Siberian high and its relations to cold surge in East Asia. Takaya, K., and H. Nakamura, 2005: Mechanisms of intraseasonal Meteor. Atmos. Phys., 44, 281–292, doi:10.1007/BF01026822. amplification of the cold Siberian high. J. Atmos. Sci., 62, ECMWF, 2012; ERA Interim, daily. European Centre for 4423–4440, doi:10.1175/JAS3629.1. Medium-Range Weather Forecasts. Subset used: 1 November Thompson, D. W. J., and J. M. Wallace, 1998: The Arctic Oscillation 1979–28 February 2011, accessed 1 July 2012. [Available on- signature in the wintertime geopotential height and temperature line at http://apps.ecmwf.int/datasets/data/interim-full-daily/.] fields. Geophys. Res. Lett., 25, 1297–1300, doi:10.1029/98GL00950. Gong, D.-Y., and C.-H. Ho, 2004: Intra-seasonal variability of ——, and ——, 2001: Regional climate impacts of the Northern wintertime temperature over East Asia. Int. J. Climatol., 24, Hemisphere annular mode. Science, 293, 85–89, doi:10.1126/ 131–144, doi:10.1002/joc.1006. science.1058958. Hayasaki, M., S. Sugata, and H. L. Tanaka, 2006: Interannual ——, M. P. Baldwin, and J. M. Wallace, 2002: Stratospheric con- variation of cold frontal activity in spring in Mongolia. nection to Northern Hemisphere wintertime weather: Impli- J. Meteor. Soc. Japan, 84, 463–475, doi:10.2151/jmsj.84.463. cations for prediction. J. Climate, 15, 1421–1428, doi:10.1175/ Hurrell, J. W., and H. van Loon, 1997: Decadal variations in cli- 1520-0442(2002)015,1421:SCTNHW.2.0.CO;2. mate associated with the North Atlantic Oscillation. Climatic Wallace, J. M., 2000: North Atlantic Oscillation/annular mode: Change, 36, 301–326, doi:10.1023/A:1005314315270. Two paradigms—One phenomenon. Quart. J. Roy. Meteor. Iwasaki, T., and Y. Mochizuki, 2012: Mass-weighted isentropic Soc., 126, 791–805, doi:10.1256/smsqj.56401. zonal mean equatorward flow in the Northern Hemispheric Walsh, J. E., A. S. Phillips, D. H. Portis, and W. L. Chapman, 2001: winter. SOLA, 8, 115–118, doi:10.2151/sola.2012-029. Extreme cold outbreaks in the United States and Europe, 1948–99. ——, T. Shoji, Y. Kanno, M. Sawada, K. Takaya, and M. Ujiie, J. Climate, 14, 2642–2658, doi:10.1175/1520-0442(2001)014,2642: 2014: Isentropic analysis of polar cold air mass streams in the ECOITU.2.0.CO;2. Northern Hemispheric winter. J. Atmos. Sci., 71, 2230–2243, Wettstein, J. J., and L. O. Mearns, 2002: The influence of the North doi:10.1175/JAS-D-13-058.1. Atlantic–Arctic Oscillation on mean, variance, and extremes of Johnson, D. R., 1989: The forcing and maintenance of global temperature in the northeastern United States and Canada. monsoonal circulations: An isentropic analysis. Adv. Geo- J. Climate, 15, 3586–3600, doi:10.1175/1520-0442(2002)015,3586: phys., 31, 43–316, doi:10.1016/S0065-2687(08)60053-9. TIOTNA.2.0.CO;2. Kolstad, E. W., T. Breiteig, and A. A. Scaife, 2010: The association Wexler, H., 1951: Anticyclones. Compendium of Meteorology, between stratospheric weak polar vortex events and cold air T. F. Malone, Ed., Amer. Meteor. Soc., 621–628. outbreaks in the Northern Hemisphere. Quart. J. Roy. Meteor. Wu, M.-C., and J. C.-L. Chan, 1995: Surface features of winter mon- Soc., 136, 886–893, doi:10.1002/qj.620. soon surges over south China. Mon. Wea. Rev., 123, 662–680, Konrad, C. E., 1996: Relationships between the intensity of cold-air doi:10.1175/1520-0493(1995)123,0662:SFOWMS.2.0.CO;2. outbreaks and the evolution of synoptic and planetary-scale Yu, Y.-Y., R.-C. Ren, J.-G. Hu, and G.-X. Wu, 2014: A mass features over North America. Mon. Wea. Rev., 124, 1067–1083, budget analysis on the interannual variability of the polar doi:10.1175/1520-0493(1996)124,1067:RBTIOC.2.0.CO;2. surface pressure in the winter season. J. Atmos. Sci., 71, 3539– Luo, D.-H., Y. Yao, and S. B. Feldstein, 2014: Regime transition of 3553, doi:10.1175/JAS-D-13-0365.1. the North Atlantic Oscillation and the extreme cold event ——, M. Cai, R.-C. Ren, and H. Van den Dool, 2015: Relationship over Europe in January–February 2012. Mon. Wea. Rev., 142, between warm airmass transport into upper polar atmosphere 4735–4757, doi:10.1175/MWR-D-13-00234.1. and cold air outbreaks in winter. J. Atmos. Sci., 72, 349–368, Namias, J., 1950: The index cycle and its role in the general circulation. doi:10.1175/JAS-D-14-0111.1. J. Meteor., 7, 130–139, doi:10.1175/1520-0469(1950)007,0130: Zhang, Q., C.-S. Shin, H. Van den Dool, and M. Cai, 2013: CFSv2 TICAIR.2.0.CO;2. prediction skill of stratospheric temperature anomalies. Cli- Palmer, T., 2014: Record-breaking winters and global climate mate Dyn., 41, 2231–2249, doi:10.1007/s00382-013-1907-5. change. Science, 344, 803–804, doi:10.1126/science.1255147. Zhang, Y., K. R. Sperber, and J. S. Boyle, 1997: Climatology and in- Pauluis, O., A. Czaja, and R. Korty, 2008: The global atmospheric terannual variation of the East Asian winter monsoon: Results from circulation on moist isentropes. Science, 321, 1075–1078, the 1979–95 NCEP/NCAR reanalysis. Mon. Wea. Rev., 125, 2605– doi:10.1126/science.1159649. 2619, doi:10.1175/1520-0493(1997)125,2605:CAIVOT.2.0.CO;2.

Unauthenticated | Downloaded 09/30/21 11:46 PM UTC