on the influence of north pacific sea surface temperature on the arctic winter climate

On the influence of North Pacific sea surfacetemperature on the Arctic winter climate

M. M. Hurwitz,1,2 P. A. Newman,2 and C. I. Garfinkel3

Received 22 March 2012; revised 24 August 2012; accepted 31 August 2012; published 5 October 2012.

[1] Differences between two ensembles of Goddard Earth Observing SystemChemistry-Climate Model simulations isolate the impact of North Pacific sea surfacetemperatures (SSTs) on the Arctic winter climate. One ensemble of extended winterseason forecasts is forced by unusually high SSTs in the North Pacific, while in the secondensemble SSTs in the North Pacific are unusually low. High – Low differences areconsistent with a strengthened Western Pacific atmospheric teleconnection pattern, and inparticular, a weakening of the Aleutian low. This relative change in tropospheric circulationinhibits planetary wave propagation into the stratosphere, in turn reducing polarstratospheric temperature in mid- and late winter. The number of winters with suddenstratospheric warmings is approximately tripled in the Low ensemble as compared with theHigh ensemble. Enhanced North Pacific SSTs, and thus a more stable and persistentArctic vortex, lead to a relative decrease in lower stratospheric ozone in spring,affecting the April clear-sky UV index at Northern Hemisphere midlatitudes.

Citation: Hurwitz, M. M., P. A. Newman, and C. I. Garfinkel (2012), On the influence of North Pacific sea surface temperatureon the Arctic winter climate, J. Geophys. Res., 117, D19110, doi:10.1029/2012JD017819.

1. Introduction

[2] The severity of Arctic ozone depletion is highlydependent on the evolution of polar lower stratospherictemperature in late winter and spring [World MeteorologicalOrganization, 2011]. In 2011, unprecedented Arctic ozonedepletion resulted from a sustained period of below-averagetemperatures and an extremely isolated polar air mass[Manney et al., 2011]. The cold 2010–2011 winter resultedin a large volume of polar stratospheric clouds (PSCs), andthereby, high levels of activated chlorine that catalyticallydestroyed ozone in the Arctic lower stratosphere [Manneyet al., 2011]. It is important to understand the causes ofthese 2011 conditions, in order to assess how Arctic ozoneand ultraviolet (UV) radiation levels will likely evolve asthe abundance of ozone-depleting substances declines.[3] Hurwitz et al. [2011a] considered possible dynamical

causes of the 2011 ozone depletion event. March was thefocus of the authors’ study, because conditions in the Arcticstratosphere were particularly anomalous during that month.Several causes were rejected: Direct radiative cooling bygreenhouse gases was dismissed because cooling of the

Arctic lower stratosphere observed during the satellite era(0.17 � 0.14 K year�1, in the MERRA reanalysis) was tooweak to explain the roughly 10 K cooling in 2011. Both ElNiño/Southern Oscillation (ENSO) and the quasi-biennialoscillation (QBO) modulate the strength of the Arctic vortexbut could not explain the anomalous stratospheric cooling in2011. Relative strengthening of the Arctic vortex during LaNiña events, as observed in 2011, weakens and begins toreverse by March. Similarly, the authors found that relativestrengthening of the mid-winter vortex during the westerlyphase of the QBO, relatively to the easterly phase, does notpersist through March. Furthermore, the structure and mag-nitude of dynamical anomalies in the Arctic stratospherewere similar in March 1997 and March 2011, two episodesof low temperatures and large ozone losses, despite differentphases of the QBO.[4] Hurwitz et al. [2011a] showed that winters with excep-

tionally high North Pacific SSTs, including 2010–2011, wereoften characterized by weak planetary wave driving in mid-winter and by cold, persistent Arctic vortices, providing theconditions necessary for severe polar ozone loss. Similarly,Jadin et al. [2010] found positive correlations between mid-winter North Pacific SSTs and the strength of the Arcticvortex. North Pacific SSTs may affect planetary wave drivingby modifying the large-scale circulation. The West Pacific(WP) teleconnection pattern is characterized by positiveupper tropospheric ridges in the western and central NorthPacific [Wallace and Gutzler, 1981]. Strong, positive WPpattern events tend to inhibit planetary wave driving andstrengthen the Arctic vortex in winter [Garfinkel andHartmann, 2008; Orsolini et al., 2009; Woollings et al.,2010; Garfinkel et al., 2010]. North Pacific ridges are an

1Goddard Earth Sciences Technology and Research, Morgan StateUniversity, Baltimore, Maryland, USA.

2NASA Goddard Space Flight Center, Greenbelt, Maryland, USA.3Earth and Planetary Science, Johns Hopkins University, Baltimore,

Maryland, USA.

Corresponding author: M. M. Hurwitz, NASA Goddard SpaceFlight Center, Code 614, Greenbelt, MD 20771, USA.([email protected])

©2012. American Geophysical Union. All Rights Reserved.0148-0227/12/2012JD017819

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 117, D19110, doi:10.1029/2012JD017819, 2012

D19110 1 of 13

efficient way to reduce planetary wave driving in theNorthern Hemisphere because they destructively interferewith the climatological stationary wave pattern [Garfinkeland Hartmann, 2008; Orsolini et al., 2009; Nishii et al.,2010; Garfinkel et al., 2010; Nishii et al., 2011]. Specifi-cally, ridges in the North Pacific (i.e., strengthening of theWP pattern) have been shown to precede prolonged polarstratospheric cooling of up to 6 K [Nishii et al., 2010] andmonths with very high PSC volumes [Orsolini et al., 2009].[5] While observational evidence suggests that North

Pacific SSTs can affect polar stratospheric conditions, theArctic winter climate is highly variable and few anomalouscooling events have occurred during the satellite era. Thus, itis not possible to attribute the dynamical cause(s) of a par-ticular event, such as the unusual meteorology and ozone lossobserved in 2011. To pinpoint the role of North Pacific SSTson the Arctic troposphere and stratosphere in winter, thepresent study compares two ensembles of chemistry-climatemodel (CCM) simulations forced by composites of observedSSTs, each providing many samples of the atmosphericresponse to high or low North Pacific SSTs. Section 2describes the model and experimental set-up. Section 3 willdiagnose ensemble mean differences in Arctic geopotentialheight, eddy heat flux, temperature, ozone and UV index.Also, Section 3 will assess the frequency of strong eddy heatflux events and major sudden stratospheric warmings(SSWs), characterized by a reversal of the climatologicalwesterly zonal winds at 60�N, 10 hPa, in the two ensembles.Section 4 will provide a discussion and summary.

2. Method

[6] The Arctic winter response to changes in North PacificSSTs is simulated using the Goddard Earth Observing SystemChemistry-Climate Model, Version 2 (GEOS V2 CCM). TheGEOS V2 CCM couples the GEOS-5 general circulationmodel (GCM) with a comprehensive stratospheric chemistrymodule [Bloom et al., 2005; Pawson et al., 2008]. The modelhas 2� latitude � 2.5� longitude horizontal resolution and 72vertical layers, with a model top at 0.01 hPa. Predicted dis-tributions of water vapor, ozone, greenhouse gases (CO2,CH4, and N2O) and CFCs (CFC-11 and CFC-12) feed back tothe radiative calculations. An earlier version of the GEOS V2CCM generally performed well in the SPARC CCMVal[2010] detailed evaluation of stratospheric processes, thoughin the Arctic winter, lower stratospheric temperatures werewarm-biased and thus the model tended to underestimatepolar ozone loss [see SPARC CCMVal, 2010, Figures 4.1 and6.37]. The present formulation of the GEOS V2 CCM is asdescribed by Hurwitz et al. [2011b]. The model formulationincludes an updated GCM, with an improved representationof tropical stationary wave patterns, and a new gravity wavedrag scheme that allows the model to produce an internally-generated quasi-biennial oscillation (QBO) with realisticperiodicity and magnitude.[7] This study compares two GEOS V2 CCM ensembles,

each composed of 40 simulations of an extended Arcticwinter season. Each of the 40 pairs of simulations is initial-ized independently, using a 1st October restart file from aperpetual ENSO neutral (ENSON) simulation [Hurwitz et al.,2011b], and is run through April 30th. The independent set ofinitial conditions, together with the model’s internal QBO,

imply that the simulated winter conditions sample the rangeof possible QBO phases. Both ensembles are forced bygreenhouse gas and ozone-depleting substances representa-tive of the 2005 climate (as in Hurwitz et al. [2011b]).[8] The two ensembles differ only by the SST and sea ice

boundary conditions imposed north of 20�N. Annuallyrepeating SST and sea ice climatologies are each constructedfrom the average of two observed winters when North Pacific(40–50�N, 160–200�E) SSTs were either unusually high(1990–1991 and 1996–1997; Figure 1a, blue stars) orunusually low (1986–1987 and 1987–1988; Figure 1a, cyanstars). Note that this experimental design tests the atmo-spheric sensitivity to forced changes of North Pacific SSTs,rather than the full non-linear interaction of the ocean–atmosphere system. South of 20�N, ENSO neutral SST andsea ice climatologies are prescribed in both ensembles (as inthe ENSON simulation, described byHurwitz et al. [2011b]).Prescribing ENSO neutral conditions eliminates the impactsof El Niño and La Niña conditions on the simulated Arcticstratospheric response. SST and sea ice values are obtainedfrom the HadISST1 data set [Rayner et al., 2003].[9] Figure 1b shows the evolution of North Pacific SST

differences throughout the winter. High – Low differencesincrease from 1 K in October to 2 K in January, and decreaseback to 1 K by the end of each simulation. The seasonalevolution of the High – Low SST differences resembles thatof the winter 2010–2011 North Pacific SST anomalies,though the magnitude of the imposed High – Low SST dif-ferences is approximately twice as large. Figure 1c showsJanuary/February SSTs used as boundary conditions in theHigh ensemble. In the North Pacific, SSTs range from 5 to8�C. Figure 1d shows January/February High – Low SSTdifferences. Positive High – Low differences span the Pacificat midlatitudes, with negative SST differences along the westcoast of North America and Alaska, and in the subtropicalwestern Pacific. Note that the pattern of High – Low SSTdifferences resembles the pattern of SST anomalies observedduring the 2010–2011 winter (not shown).[10] In order to interpret the High – Low differences,

simulated stratospheric ozone and temperature fields arecompared with observational data sets. Simulated totalozone is compared with the TOMS/SBUV data set (updatedfrom Stolarski and Frith [2006]). Simulated temperaturefields are compared with the Modern Era Retrospective-Analysis for Research and Applications (MERRA) reanaly-sis. The MERRA reanalysis is based on an extensive set ofsatellite observations and on the Goddard Earth ObservingSystem Data Analysis System, Version 5 (GEOS-5)[Bosilovich, 2008; Rienecker et al., 2011]. The MERRAreanalysis has vertical coverage up to 0.1 hPa, and for thisstudy, is interpolated to 1.25� � 1.25� horizontal resolution.Both the TOMS/SBUV and MERRA data sets span the1979–2011 period.

3. Differences Between the Highand Low Ensembles

3.1. Troposphere

[11] Figure 2 shows High – Low monthly mean geopo-tential differences at Arctic latitudes in December andMarch. At 850 hPa, a region of positive geopotential heightdifferences in the North Pacific, co-located with the region

HURWITZ ET AL.: NORTH PACIFIC SSTs AND ARCTIC CLIMATE D19110D19110

2 of 13

of increased SSTs (see Figure 1d), develops in December(Figure 2a) and persists through the spring (Figure 2d).Positive geopotential height differences are also co-locatedwith the climatological trough in the North Pacific (gray,dashed contours), suggesting that a relative increase in NorthPacific SSTs tends to decrease tropospheric wave driving. Aregion of negative differences develops over the Arctic capin December (Figure 2a), and persists through the spring(Figure 2d). In January and February, sea level pressuredecreases at polar latitudes, and increases in the Europeanand Atlantic sectors (Figure 3). This pattern of sea levelpressure differences is consistent with a positive shift in theNorth Atlantic Oscillation (NAO). Specifically, the sea levelpressure difference between 334�E, 38�N (Ponta Delgada,Azores) and 338�E, 64�N (Reykjavik, Iceland) is signifi-cantly different in the High and Low simulations at 95%confidence level in January.[12] The tropospheric response to the North Pacific SST

difference is structurally barotropic. Consistent withFrankignoul and Sennéchael [2007], upper tropospheric(300 hPa; Figures 2b and 2e) geopotential height differences

mimic the near-surface (850 hPa; Figures 2a and 2d) geo-potential height differences. From January through March,High – Low differences indicate a strengthening of theArctic Oscillation: negative differences at Arctic latitudes,and positive differences at midlatitudes, particularly in thePacific and Atlantic sectors.[13] Eddy heat flux at 40–80�N, 100 hPa is a measure of

the planetary wave activity entering the Arctic stratosphere.Time-integrated eddy heat flux in this region is highly cor-related with polar lower stratospheric temperatures, with a1–2 month lag [Newman et al., 2001]. Figure 4 shows themean eddy heat flux in the preceding 45 days, in the High(blue (Figure 4, top)) and Low (cyan) ensembles, as well asthe mean differences between the two ensembles (dottedblack line (Figure 4, bottom)). High – Low differences arenegative from early January through mid-February, sug-gesting a relative High – Low decrease in eddy heat flux inDecember and January; in Figure 4, statistical significanceof mean differences is based on a two-tailed t-test. By earlyFebruary, eddy heat flux in the Low ensemble is both larger

Figure 1. (a) Time series of the January/February mean North Pacific SST index. Blue (cyan) starsdenote winters used to construct the High (Low) SST and sea ice boundary conditions. The red stardenotes the North Pacific SST index in 2011. (b) Time series of North Pacific SST anomalies throughthe extended winter season, for High – Low (black line) and 2010–2011 anomalies from the 1979–2011 climatology (red line). (c) January/February mean SSTs used as boundary conditions in the Highensemble. (d) January/February mean differences between the High and Low SST fields. White contoursindicate zero difference. Black boxes in Figures 1c and 1d indicate the North Pacific region.


3 of 13

and more variable than in the High ensemble; the signifi-cance of the difference in variability is based on an F-test.The sign of High – Low eddy heat flux differences flips inmid-April (Figure 4, bottom).[14] Not only do the 45-day mean eddy heat fluxes

differ between the two ensembles, so does the frequencyof strong wave events. Table 1 lists the total of dayswhere the eddy heat flux at 40–80�N, 100 hPa is greateror equal to 25 K m s�1. Note that days when eddy heatflux is greater or equal to 25 K m s�1 represent approxi-mately 5% of the simulated period. Strong eddy heat flux“events” are defined as the persistence of 40–80�N,100 hPa eddy heat flux, greater or equal to 25 K m s�1,for two or more consecutive days. The total number ofstrong eddy heat flux events is larger in the Low ensemblethan the High ensemble, as is the number of strong eddyheat flux days. In December and February, note that thenumber of strong heat flux days is roughly doubled in theLow ensemble as compared with the High ensemble.

3.2. Polar Stratosphere

[15] Polar geopotential height at 50 hPa is a measure bothof the strength of the Arctic vortex, and of the mean tem-perature below 50 hPa. Negative High – Low geopotentialheight differences over the polar cap (Figure 2f) persist fromJanuary through March. Negative geopotential height dif-ferences shift toward the Siberian sector in April. Polar captemperature at 50 hPa is used to examine the seasonal

Figure 2. High – Low geopotential height differences (m) at (a and d) 850 hPa, (b and e) 300 hPaand (c and f) 50 hPa, in December (Figures 2a–2c) and March (Figures 2d–2f), poleward of 30�N.Note the different color scale for each pressure level. White contours indicate zero difference. Graydashed contours show mean geopotential heights in the ENSO neutral simulation. In each panel, theblack wedge indicates the North Pacific region. High – Low differences are in general statistically sig-nificant at the 95% level, in a two-tailed t-test.

Figure 3. January/February High – Low sea level pressuredifferences (hPa), poleward of 30�N. White contours indi-cate zero difference. Black Xs indicate differences signifi-cant at the 95% level, in a two-tailed t-test.


4 of 13

evolution of the High – Low changes. The polar cap cools inthe mid- to late winter, in the High ensemble relative to theLow ensemble (Figure 4, bottom). 50 hPa polar cap tem-perature differences are statistically significant in Januaryand February, consistent with the negative eddy heat fluxdifferences, and in late March (black and gray Xs belowFigure 4 (center)). Note that approximately 2/3 of the 40pairs of simulated winters show a strong High – Lowcooling of the polar stratosphere in March. Polar captemperature differences reverse sign in mid-April (similarto the seasonal evolution of total ozone in 2010 and 2011[Balis et al., 2011]). Early winter temperature differencesare negligible.[16] North Pacific SSTs modulate not only the mean state

of the polar stratosphere, but also the distribution of possiblewinter conditions. In Figure 4 (middle), blue (cyan) dashedlines indicate the 90th-percentile (fourth-highest) and 10th-percentile (fourth-lowest) temperature values of the High(Low) ensemble. The 10th-percentile values are similar in

the two ensembles. However, the 90th-percentile values aresignificantly warmer (denoted by the Xs above the plot) inthe Low ensemble than in the High ensemble in January andearly February. This suggests that mid-winter polar strato-spheric variability is enhanced when North Pacific SSTs arerelatively lower. Section 3.3 will relate this result to the

Figure 4. (top) Eddy heat flux (K m s�1) at 40–80�N, 100 hPa in the preceding 45 days (i.e., cumulativeeddy heat flux). Solid blue and cyan lines indicate the High and Low ensemble mean values. Dashed blueand cyan lines indicate the 10th- and 90th-percentile values for each ensemble. Black Xs below the plotindicate days when the High – Low eddy heat flux difference is significant at the 95% level, in a two-tailedt-test. Black Xs above the plot indicate days when the difference in variability between the High and Lowensembles is significant at the 95% level, in an F-test. (middle) Like Figure 4 (top) but for polar cap tem-perature (K) at 50 hPa. (bottom) High – Low ensemble mean differences in eddy heat flux (dotted blackline) and temperature (solid black line).

Table 1. Number of Days When Eddy Heat Flux at 40–80�N,100 hPa is Equal to or Greater Than 25 K m s�1, in December,January and February, in the High and Low Ensemblesa

Number of Events Total

Number of Days

Total Dec Jan Feb

High 143 414 89 82 48Low 153 497 138 91 103

aEvents are defined as the persistence of 40–80�N, 100 hPa eddy heatflux, greater or equal to 25 K m s�1, for two or more consecutive days.Totals represent the November–March extended winter season.


5 of 13

relative frequency of sudden stratospheric warmings (SSWs)in the two ensembles.[17] The mean and distribution of seasonal mean January–

February–March (JFM) polar cap temperature at 50 hPa dif-fers robustly between the two ensembles. Figure 5a showshistograms of the JFM mean polar cap temperature; the meanHigh – Low difference is 1.91 K. Note that temperatures in theLow ensemble (cyan bars) have a tail toward high values.[18] In contrast, the JFM mean polar cap 50 hPa temper-

ature is independent of the modeled QBO phase. Figure 5bshows histograms of the JFM polar cap temperature in

both the High and Low ensembles, partitioned by QBOphase (i.e., the magnitude of JFM zonal winds at 30 hPa,between 10�S and 10�N, following Hurwitz et al. [2011b]).QBO-westerly years are characterized by JFM zonal windsgreater than 2 m s�1, while QBO-easterly years are charac-terized by zonal winds less than �2 m s�1. The means of theQBO-W and QBO-E composites are statistically indistin-guishable, in a two-tailed t-test. In their observational anal-ysis of polar stratospheric conditions in March,Hurwitz et al.[2011a] found a similar lack of dependence on the QBOphase. Note that the QBO does modulate the strength of the

Figure 5. Histograms of the January-February-March (JFM) seasonal mean polar cap temperature (K) at50 hPa: (a) High and Low ensembles, and (b) QBO-easterly years and QBO-westerly years. The black-tipped bars indicate the location of the ensemble mean values; the High and Low ensemble mean differ-ence is significant at the 95% level, in a two-tailed t-test.


6 of 13

Arctic stratosphere in early winter (i.e., the Holton-Tanrelation [Holton and Tan, 1980]), in this model version. Alsonote the relative abundance of simulated QBO-E years, ascompared with QBO-W.[19] Polar stratospheric temperature anomalies descend

from the upper stratosphere in mid-winter to the lowerstratosphere in spring. The evolution of High – Low tem-peratures at 80�N through the extended winter season issummarized by Figure 6. Temperature differences are neg-ligible in November through mid-December. By lateDecember, relative cooling develops at and above 10 hPa. InJanuary, relative cooling of the middle stratosphere exceeds8 K. The anomalous cooling slowly descends, with weakercooling extending to the 400-hPa level in February. Lowerstratospheric cooling is strongest in March and early April.In March, relative cooling of 1–2 K extends to the surfacelayer, suggestive of downward stratosphere-tropospherecoupling. Stratospheric cooling is followed by relativewarming of 2–4 K, with an approximately two-month lag.The Arctic stratospheric temperature response to ENSOevents has a similar structure [Manzini et al., 2006].[20] The breakup of the Arctic vortex, as defined by the

transition from mean westerlies to easterlies at 60�N, occurslater in the High ensemble than in the Low ensemble. At10 hPa, the mean difference in the breakup date at 10 hPa isapproximately 4 days (not shown). The relatively laterbreakup in the High ensemble is consistent with the rela-tively colder polar lower stratosphere in late winter (seeFigures 4 through 6).

[21] Interannual variability of mid-winter North PacificSSTs explains some of the observed interannual variabilityin polar lower stratospheric conditions in late winter.Figure 7 shows the observed February/March volume ofpolar stratospheric clouds (VPSC) time series for the 380–550 K potential temperature layer, updated from Rex et al.[2004, 2006], versus the January/February North Pacific

Figure 6. High – Low temperature differences (K) at 80�N, as a function of date and altitude. White con-tours indicate zero difference. Black Xs indicate differences significant at the 95% confidence level, in atwo-tailed t-test.

Figure 7. January/February North Pacific SST anomaly (K)versus February/March VPSC in the 380–550 K layer(106 km3), for each winter in the 1980–2011 period. SSTanomaly and VPSC values are denoted by year number (e.g.,‘11’ denotes 2011).


7 of 13

SST index (as in Hurwitz et al. [2011a]) in the same winter.VPSC represents the volume of polar air that is below theformation threshold for polar stratospheric clouds (approxi-mately 195 K at 50 hPa). Thus, February/March VPSC is ametric of the potential for late winter ozone depletion. TheVPSC correlation with the January/February North PacificSST index is greater for February/March (r = 0.38) than forthe December–March average (r = 0.33; not shown). Fur-thermore, the SST-VPSC correlations increase when only thelargest VPSC values are considered: r = 0.76 for February/March VPSC ≥ 15 � 106 km3; this result is consistent withthe finding that blocking in the North Pacific tends to pre-cede very high monthly mean VPSC values [Orsolini et al.,2009]. However, while VPSC and North Pacific SSTs wereboth unusually high in 2011, elevated North Pacific SSTs donot explain the near-zero VPSC in e.g., 1991.

3.3. Sudden Stratospheric Warmings

[22] Major sudden stratospheric warmings (SSWs) arecharacterized by reversals in the zonal wind direction at60�N, 10 hPa. The observed occurrence of SSW events,during years when January/February North Pacific SSTs

were unusually high or low, is compiled from a list of his-torical SSWs [Butler and Polvani, 2011]. Distinct SSWs,and thus distinct zonal wind reversals, require at least a20-day separation, and must be distinct from the finalwarming (i.e., breakup of the Arctic vortex) [Charltonand Polvani, 2007]. SSWs occurred during all of thefour winters with the lowest North Pacific SSTs between1979 and 2011 (see Figure 1a), while no major SSWswere observed during the four winters with the highestNorth Pacific SSTs. The GEOS V2 CCM ensembles testwhether or not this strong observed sensitivity of theSSW frequency to North Pacific SSTs is robust to alarger sample size.[23] Table 2 shows the number and frequency of SSWs in

the two GEOS V2 CCM ensembles. Simulated SSW iden-tification follows Charlton and Polvani [2007] and Butlerand Polvani [2011]. The relative frequency of simulatedSSWs is consistent with the mean strength of the Arcticvortex: the frequency of winters with at least one SSW (i.e.,active winters) is approximately tripled in the Low ensemble(20 active winters) as compared with the High ensemble (7active winters). There are two winters each with two SSWs

Table 2. Simulated Number and Frequency of Winters With at Least One SSW, and Number and Frequency of Winters With Two SSWsa

GEOS V2 CCM Simulations

1979–2011 Frequencyof Active Winters

Number ofActive Winters

Frequency ofActive Winters

Number of Winterswith Two SSWs

Frequency ofTwo-SSW Winters

High 7 0.18 2 0.05 0Low 20 0.50 2 0.05 1ENSO Neutral 10 0.25 0 0 0.42

aSimulated number and frequency of winters with at least one SSW: 2nd and 3rd columns; number and frequency of winters with two SSWs: 4th and 5thcolumns. Forty winters for each of the High, Low and ENSO neutral simulations are considered. The 6th column shows the frequency of active wintersduring the 1979–2011 period, based on a list of historical SSWs [Butler and Polvani, 2011]. For the historical SSWs, the four years with the highest(lowest) January/February North Pacific SSTs represent the High (Low) case.

Figure 8. Zonally asymmetric component of the 300-hPa geopotential height anomalies in the 5–20 dayspreceding SSWs in the (a) High and (b) Low ensembles, poleward of 30�N. White contours indicate zerodifference.


8 of 13

in both ensembles. The increased SSW frequency in the Lowensemble is consistent with the overall increase in variabilityin January (see Figure 4). The frequency of active wintersduring perpetual ENSO neutral conditions (10 active wintersin a 40-year sample) lies between that for the North PacificSST extremes.[24] The dynamical situation preceding SSWs in the High

ensemble is distinct from that preceding SSWs in the Lowensemble. Composites of the zonally asymmetric component

of the 300-hPa geopotential height anomalies in the 5–20 days preceding SSWs, at Arctic latitudes, highlight dif-ferences between the ensembles: The High ensemble has awave number–2-like structure (Figure 8a) while the Lowensemble has a wave number–1-like structure (Figure 8b).These structures are, respectively, consistent with the tropo-spheric precursors preceding split and displacement typeSSWs [Charlton and Polvani, 2007; Cohen and Jones,2012].

Figure 9. High – Low differences in ozone mixing ratio [ppmv] and in the residual vertical velocity (w*)(mm s�1) in the 70–90�N region, as a function of date and altitude. Filled, colored contours show w* dif-ferences; white contours indicate zero w* difference. Black contours indicate ozone differences; thick,black contours indicate zero ozone difference.

Figure 10. Time series of zonal mean temperature at 80�N, 50 hPa during the 2010–2011 winter (black),as compared with the High (blue) and Low (cyan) ensemble means. The approximate threshold for PSCformation (195 K) is denoted by the dotted black line.


9 of 13

3.4. High-Latitude Stratospheric Ozone and UV

[25] High – Low polar ozone differences are shown inFigure 9. In January through March, ozone differences in themiddle and lower stratosphere are not statistically signifi-cant, despite strong polar cooling (Figure 6) and weakenedpolar downwelling (positive w* differences; see coloredcontours, Figure 9, and Balis et al. [2011]). The reversal inthe sign of High – Low 50 hPa temperature differences inmid-April (see Figures 4 and 6) does not occur in the ozonedifference field. Weak, negative ozone differences of 0.1–0.2 ppmv, centered at 50 hPa, occur in late March and April,consistent with the relative cooling at 80�N (Figure 6).[26] Weak High – Low ozone differences result from a

polar lower stratospheric warm bias in the GEOS V2 CCM.Figure 10 shows that the High and Low ensemble meantemperatures at 80�N, 50 hPa remain above the threshold forPSC formation (approximated by the black dotted line), andthus for chemical ozone depletion, throughout the winter. Incontrast, observed temperatures during the e.g., 2010–2011winter (solid black line) dropped below the PSC formationthreshold for sustained periods in February and March.Though the simulated area where polar temperatures arebelow 195 K at 50 hPa is never large enough to match the

ozone loss observed in e.g., 2011 (see Figure 9 and Manneyet al. [2011]), the 50-hPa PSC area in the High and Lowensembles are significantly different in January, Februaryand March (see Table 3).[27] In April, High – Low ensemble mean total ozone

differences are consistent with the observed total ozoneanomalies in 2011. In the GEOS CCM simulations(Figure 11a), positive differences of 5–10 DU are seen in theCanadian sector, while negative differences of up to 37 DUare seen in the European and Siberian sectors. Weaker,negative total ozone differences span the midlatitude Pacificand the eastern USA. The observed total ozone anomalies inApril 2011 (Figure 11b) match the pattern of the simulateddifferences, but with two main differences: First, theobserved anomalies are approximately three times larger.Second, the observed response lacks the simulated negativetotal ozone response in the midlatitude East Asia and Pacificsector.[28] Decreased total ozone in spring enhances the flux of

clear-sky UV radiation to the earth’s surface, while increasedtotal ozone diminishes the UV flux. Simulated April UVdifferences are calculated for four cities: Washington,Calgary, London and Moscow. These cities span the latitudi-nal circle at northern midlatitudes. Figure 12 shows the AprilUV index in the four cities, in two years when North PacificSSTs were unusually low, three years when North PacificSSTs were unusually high, as well as the High and Lowensemble mean values. In Washington, London and Moscow,North Pacific SSTs and the April UV index are positivelycorrelated: UV index is highest when North Pacific SSTs areelevated and Arctic ozone depletion is relatively more severe.The statistical significance of High – Low UV differences ishighest for Washington. In Calgary, the UV index exhibits

Table 3. Ensemble and Monthly Mean Area Where 50-hPa PolarTemperature is Less Than 195 K (106 m2)a

Dec Jan Feb Mar

High 2.6 � 1.2 19.2 � 2.6 27.3 � 2.5 12.1 � 2.2Low 3.7 � 1.2 11.4 � 2.5 16.4 � 2.8 6.9 � 1.7

aDifferences between the High and Low areas are significantly significantat the 90% level, in a two-tailed t-test, in January, February and March.

Figure 11. April total ozone differences (DU) (a) for the High – Low ensemble means and (b) for 2011 –climatology from the TOMS/SBUV data set, poleward of 30�N. In Figure 11a, black Xs indicate differ-ences significant at the 95% level in a two-tailed t-test. White letters denote the approximate location ofthe four cities: London (L), Moscow (M), Calgary (C) and Washington (W).


10 of 13

the opposite sensitivity to North Pacific SSTs, consistentwith the total ozone response in April.

4. Conclusions

[29] GEOS V2 CCM simulations show that SSTs in theNorth Pacific region significantly affect the Arctic tropo-sphere and stratosphere in winter. Two ensembles, each of40 extended winter season simulations, differ only by theSST boundary conditions imposed north of 20�N. Thelargest, positive SST differences are in the North Pacificregion. By imposing no tropical SST differences, the impactof ENSO on Arctic winter variability is removed. While thephase of the QBO varies between each of the ensemblemembers, differences in the simulated Arctic late winterresponse during QBO-easterly versus QBO-westerly condi-tions are negligible.[30] In the troposphere, enhanced North Pacific SSTs tend

to weaken the Aleutian low (consistent with Latif andBarnett [1996], Liu et al. [2006], and Frankignoul andSennéchael [2007]), inhibiting planetary wave propagationinto the stratosphere. Arctic geopotential heights are rela-tively lower, from December through March. Furthermore,enhanced North Pacific SSTs can affect winter weather inthe Northern Hemisphere: relatively lower surface pressureat polar latitudes coupled with relatively higher pressure atmidlatitudes imply a strengthening of the North AtlanticOscillation and Arctic Oscillation (consistent with correla-tion analysis by Kim and Ahn [2012], for the autumn sea-son). However, SST differences between the High and Low

ensembles are non-zero in the North Atlantic, and thus,imposed Atlantic SST differences may have contributed tothe simulated tropospheric response.[31] Reduced tropospheric planetary wave driving, in

winters with relatively warmer North Pacific SSTs, leads toa less disturbed Arctic vortex. Lower temperatures in theArctic stratosphere begin in the upper stratosphere in Janu-ary, and slowly descend to the lower stratosphere by Marchand early April. The JFM seasonal mean difference in polarcap temperature at 50 hPa is 1.91 K. The breakup of theArctic vortex is delayed by approximately 4 days at 10 hPa.When North Pacific SSTs are relatively low, not only is theseasonal mean Arctic vortex weakened, the frequency ofsimulated SSWs is considerably larger (consistent withJadin et al. [2010]). Furthermore, the tropospheric dynami-cal situation preceding SSWs that occur when North PacificSSTs are anomalously cool (warm) tend to have a wavenumber–1 (wave number–2) structure. The reduction inSSW frequency by warmer North Pacific SSTs is consistentwith the reduction in planetary wave driving.[32] The Arctic ozone response to enhanced North Pacific

SSTs is consistent with the dynamical response. Relativecooling of the Arctic stratosphere corresponds with a relativedecrease in polar ozone. In the lower and middle strato-sphere, negative ozone differences are centered over thepolar cap from January through March. Though the patternof simulated ozone differences is consistent with thedynamical response, ozone differences are not statisticallysignificant, likely because of the GEOS V2 CCM’s mid-winter warm bias.

Figure 12. April clear-sky UV index in four cities: Washington (approximately 283�E, 38�N), Calgary(246�E, 51�N), Moscow (37�E, 55�N) and London (0�E, 51�N). For each city, the first three vertical barsrepresent the UV index calculated based on April total ozone from the TOMS/SBUV data set: the mean of1987 and 1988 (unusually low North Pacific SSTs), the mean of 1991 and 1997 (high), and 2011 (high).The last two vertical bars represent the mean UV index calculated based on April total ozone for theHigh and Low simulations. Differences in the April UV index between the High and Low ensembles aresignificant at the 95% confidence level in Washington and at the 90% confidence level in London,in two-tailed t-tests.


11 of 13

[33] In April, negative High – Low ozone differencesshift away from the pole and toward the Siberian sector.The pattern of simulated total ozone differences is similarto the pattern of ozone anomalies observed in April 2011[see also Balis et al., 2011]; however, the magnitude ofthe simulated differences is approximately a third of thatobserved. In April, negative total ozone differencesexceeding 30 DU are simulated in Europe and the Siberiansector (leading to small increases in clear-sky UV index ine.g., London and Moscow), while total ozone increasesexceed 20 DU over northern Canada (slightly decreasingthe UV index in e.g., Calgary). Weaker negative differ-ences are seen at northern midlatitudes (increasing theApril UV index in e.g., Washington).[34] The results of the present study suggest that one of

the keys to predicting the behavior of the Arctic strato-sphere in a future climate is to understand variability andtrends in the North Pacific Ocean. Interannual variability inNorth Pacific SSTs is primarily driven by a “subarcticmode” of decadal variability that is independent of ENSOand the Pacific Decadal Oscillation (PDO) [Nakamura et al.,1997; Frankignoul and Sennéchael, 2007; Orsolini et al.,2009; Nishii et al., 2010]. The Coupled Model Intercom-parison Project Phase 5 (CMIP5; http://cmip-pcmdi.llnl.gov/cmip5/) simulations of likely 21st century climate scenarioswill provide updated forecasts of future variability and trendsin North Pacific SSTs.[35] The above model results support the hypothesis that

anomalously warm North Pacific SSTs contributed to thestrong and prolonged cooling of the polar lower stratosphere,and severe ozone depletion, that were observed during the1996–1997 and 2010–2011 winters [Pawson and Naujokat,1999; Manney et al., 2011; Hurwitz et al., 2011a]. January/February North Pacific SSTs are well correlated with latewinter VPSC, explaining some of the “coldest winters” asidentified by Rex et al. [2004, 2006] and Manney et al.[2011]. However, elevated North Pacific SSTs alone do notpredict polar stratospheric cooling in late winter: Thewarmest North Pacific SSTs of the satellite era were observedduring the 1990–1991 winter, while VPSC was near-zero.Similarly, in the GEOS V2 CCM simulations, not all wintersforced by high North Pacific SSTs lead to late winter strato-spheric cooling, because stochastic atmospheric variabilityalso plays a role in modulating the Arctic winter climate.

[36] Acknowledgments. The authors thank S. Frith for providing theupdated TOMS/SBUV data and processing the model output, M. Rex andP. von der Gathen for supplying VPSC time series, A. Karpechko and threeanonymous reviewers for helpful comments, and NASA’s ACMAP pro-gram for funding.

ReferencesBalis, D., et al. (2011), Observed and modelled record ozone decline overthe Arctic during winter/spring 2011, Geophys. Res. Lett., 38, L23801,doi:10.1029/2011GL049259.

Bloom, S., et al. (2005), Documentation and validation of the GoddardEarth-Observing System (GEOS) Data Assimilation System Version 4,NASA Tech. Rep., NASA TR-104606, vol. 26, 187 pp.

Bosilovich, M. (2008), NASA’s modern era retrospective-analysis forresearch and applications: Integrating Earth observations [online], Earth-zine, 26 September. [Available at http://www.earthzine.org/2008/09/26/nasas-modern-era-retrospective-analysis/]

Butler, A. H., and L. M. Polvani (2011), El Niño, La Niña, and stratosphericsudden warmings: A reevaluation in light of the observational record,Geophys. Res. Lett., 38, L13807, doi:10.1029/2011GL048084.

Charlton, A., and L. Polvani (2007), A new look at stratospheric suddenwarmings. Part I: Climatology and modeling benchmarks, J. Clim., 20,449–469, doi:10.1175/JCLI3996.1.

Cohen, J., and J. Jones (2012), Tropospheric precursors and stratosphericwarmings, J. Clim., 24, 6562–6572, doi:10.1175/2011JCLI4160.1.

Frankignoul, C., and N. Sennéchael (2007), Observed influence of NorthPacific SST anomalies on the atmospheric circulation, J. Clim., 20,592–606, doi:10.1175/JCLI4021.1.

Garfinkel, C. I., and D. L. Hartmann (2008), Different ENSO teleconnec-tions and their effects on the stratospheric polar vortex, J. Geophys.Res., 113, D18114, doi:10.1029/2008JD009920.

Garfinkel, C. I., D. L. Hartmann, and F. Sassi (2010), Tropospheric precur-sors of anomalous Northern Hemisphere stratospheric polar vortices, J.Clim., 23, 3282–3299.

Holton, J. R., and H.-C. Tan (1980), The influence of the equatorial quasi-biennial oscillation on the global circulation at 50 mb, J. Atmos. Sci., 37,2200–2208, doi:10.1175/1520-0469(1980)037<2200:TIOTEQ>2.0.CO;2.

Hurwitz, M. M., P. A. Newman, and C. I. Garfinkel (2011a), The Arcticvortex in March 2011: A dynamical perspective, Atmos. Chem. Phys.,11, 11,447–11,453, doi:10.5194/acp-11-11447-2011.

Hurwitz, M. M., I.-S. Song, L. D. Oman, P. A. Newman, A. M. Molod,S. M. Frith, and J. E. Nielsen (2011b), Response of the Antarcticstratosphere to warm pool El Niño events in the GEOS CCM, Atmos.Chem. Phys., 11, 9659–9669, doi:10.5194/acp-11-9659-2011.

Jadin, E. A., R. Wei, Y. A. Zyalyaeva, W. Chen, and L. Wang (2010),Stratospheric wave activity and the Pacific Decadal Oscillation, J. Atmos.Solar Terr. Phys., 72, 1163–1170, doi:10.1016/j.jastp.2010.07.009.

Kim, H.-J., and J.-B. Ahn (2012), Possible impact of the autumnal NorthPacific SST and November AO on the East Asian winter temperature,J. Geophys. Res., 117, D12104, doi:10.1029/2012JD017527.

Latif, M., and T. P. Barnett (1996), Decadal climate variability over the NorthPacific and North America: Dynamics and predictability, J. Clim., 9,2407–2423, doi:10.1175/1520-0442(1996)009<2407:DCVOTN>2.0.CO;2.

Liu, Q., N. Wen, and Z. Liu (2006), An observational study of the impact ofthe North Pacific SST on the atmosphere, Geophys. Res. Lett., 33,L18611, doi:10.1029/2006GL026082.

Manney, G. L., et al. (2011), Unprecedented Arctic ozone loss in 2011,Nature, 478, 469–475, doi:10.1038/nature10556.

Manzini, E., M. A. Giorgetta, M. Esch, L. Kornblueh, and E. Roeckner(2006), The influence of sea surface temperatures on the northern winterstratosphere: Ensemble simulations with the MAECHAM5 model, J.Clim., 19, 3863–3881, doi:10.1175/JCLI3826.1.

Nakamura, H., G. Lin, and T. Yamagata (1997), Decadal climate variabilityin the North Pacific during the recent decades, Bull. Am. Meteorol. Soc.,78, 2215–2225, doi:10.1175/1520-0477(1997)078<2215:DCVITN>2.0.CO;2.

Newman, P. A., E. R. Nash, and J. E. Rosenfield (2001), What controls thetemperature of the Arctic stratosphere during the spring?, J. Geophys.Res., 106, 19,999–20,010, doi:10.1029/2000JD000061.

Nishii, K., H. Nakamura, and Y. J. Orsolini (2010), Cooling of the winter-time Arctic stratosphere induced by the western Pacific teleconnectionpattern, Geophys. Res. Lett., 37, L13805, doi:10.1029/2010GL043551.

Nishii, K., H. Nakamura, and Y. J. Orsolini (2011), Geographical depen-dence observed in blocking high influence on the stratospheric variabilitythrough enhancement and suppression of upward planetary-wave propa-gation, J. Clim., 24, 6408–6423, doi:10.1175/JCLI-D-10-05021.1.

Orsolini, Y. J., A. Y. Karpechko, and G. Nikulin (2009), Variability of theNorthern Hemisphere polar stratospheric cloud potential: The role ofNorth Pacific disturbances, Q. J. R. Meteorol. Soc., 135, 1020–1029,doi:10.1002/qj.409.

Pawson, S., and B. Naujokat (1999), The cold winters of the middle 1990sin the northern lower stratosphere, J. Geophys. Res., 104, 14,209–14,222,doi:10.1029/1999JD900211.

Pawson, S., R. S. Stolarski, A. R. Douglass, P. A. Newman, J. E. Nielsen,S. M. Frith, and M. L. Gupta (2008), Goddard Earth Observing Systemchemistry-climate model simulations of stratospheric ozone-temperaturecoupling between 1950 and 2005, J. Geophys. Res., 113, D12103,doi:10.1029/2007JD009511.

Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander,D. P. Rowell, and A. Kaplan (2003), Global analyses of sea surfacetemperature, sea ice, and night marine air temperature since the latenineteenth century, J. Geophys. Res., 108(D14), 4407, doi:10.1029/2002JD002670.

Rex, M., R. J. Salawitch, P. von der Gathen, N. R. P. Harris, M. P.Chipperfield, and B. Naujokat (2004), Arctic ozone loss and climatechange, Geophys. Res. Lett., 31, L04116, doi:10.1029/2003GL018844.

Rex, M., et al. (2006), Arctic winter 2005: Implications for stratosphericozone loss and climate change, Geophys. Res. Lett., 33, L23808,doi:10.1029/2006GL026731.


12 of 13

Rienecker, M.M., et al. (2011), MERRA—NASA’s Modern-Era Retrospec-tive Analysis for Research and Applications, J. Clim., 24, 3624–3648,doi:10.1175/JCLI-D-11-00015.1.

SPARC CCMVal (2010), SPARC report on the evaluation of chemistry-climate models, edited by V. Eyring, T. G. Shepherd, and D. W.Waugh, SPARC Rep. 5, Univ. of Toronto, Toronto, Ont., Canada.[Available at http://www.atmosp.physics.utoronto.ca/SPARC.]

Stolarski, R. S., and S. M. Frith (2006), Search for evidence of trend slow-down in the long-term TOMS/SBUV total ozone data record: The impor-tance of instrument drift uncertainty, Atmos. Chem. Phys., 6, 4057–4065,doi:10.5194/acp-6-4057-2006.

Wallace, J. M., and D. S. Gutzler (1981), Teleconnections in the geopotentialheight field during the Northern Hemisphere winter, Mon. WeatherRev., 109, 784–812, doi:10.1175/1520-0493(1981)109<0784:TITGHF>2.0.CO;2.

Woollings, T., A. Charlton–Perez, S. Ineson, A. G. Marshall, and G. Masato(2010), Associations between stratospheric variability and troposphericblocking, J. Geophys. Res., 115, D06108, doi:10.1029/2009JD012742.

World Meteorological Organization (2011), Scientific assessment of ozonedepletion: 2010, Global Ozone Res. Monit. Proj. Rep. 52, 516 pp., WMO,Geneva, Switzerland.


13 of 13

on the influence of north pacific sea surface temperature on the arctic winter climate

Documents