1. Introduction
Assessing and distinguishing between natural and anthropogenic forcing of climate is a formidable challenge (Solomon et al. 2011). To better predict how climate will vary in the future, it is useful to investigate natural modes of variability in the climate system. One of these modes operating on decadal time scales is the Pacific decadal oscillation (PDO). The PDO, which is the leading mode of nonseasonal sea surface temperature (SST) variability in the North Pacific (NP) Ocean (Mantua et al. 1997), explains approximately 25% of the nonseasonal monthly mean variance in the NP and is comparable to that explained by El Niño–Southern Oscillation (ENSO) (Deser et al. 2010). The PDO oscillates on decadal time scales between a warm and cold phase, with observable shifts since the late eighteenth century (Minobe 1997). Furthermore, since 1998, the rise in global surface temperatures has stalled, leading to a so-called hiatus in global warming (Trenberth and Fasullo 2013). It is speculated that the PDO may partly explain this hiatus and be important for decadal climate prediction.
A large body of research since the 1990s has been devoted to understanding the forcing of the PDO. Miller and Schneider (2000), for example, discuss six proposed mechanisms for decadal SST variability in the NP. Each one is not necessarily independent of the others; the forcing is likely a combination of several mechanisms acting together. These mechanisms include stochastic atmospheric forcing from variations in the Aleutian low (Jin 1997; Frankignoul et al. 1997; Seager et al. 2001), forcing from ENSO (e.g., Miller et al. 1994; Pierce 2001; Newman et al. 2003; Deser et al. 2004; Schneider and Cornuelle 2005; Vimont 2005; Chhak et al. 2009; Shakun and Shaman 2009; Kwon et al. 2010), midlatitude ocean–atmosphere interactions (Latif and Barnett 1994, 1996; Jin 1997; Bond and Harrison 2000; Seager et al. 2001; Kwon and Deser 2007; Frankignoul and Sennéchael 2007), tropical–extratropical interactions (Gu and Philander 1997; Lu et al. 1998; White and Cayan 1998; Kleeman et al. 1999; Johnson 2003; McPhaden and Zhang 2002; Molinari et al. 2003; Alexander et al. 2006), oceanic teleconnections to the midlatitudes from the tropics (Miller and Schneider 2000), and intrinsic ocean variability. For a thorough review of the forcing of the PDO, we refer the reader to Miller and Schneider (2000) and Alexander (2010) and the references therein.
The influence of the PDO on surface climate has been documented in several studies (e.g., Mantua et al. 1997; Minobe 1997; Zhang et al. 1997; Mantua and Hare 2002; Alexander et al. 2006; Deser et al. 2012). The positive phase of the PDO exhibits a horseshoe-shaped pattern in SST anomalies (Fig. 1), with cold anomalies in the western and central NP, and opposite warm anomalies in the Gulf of Alaska, off the coast of the western United States, and over the eastern tropical Pacific (Mantua et al. 1997). The positive PDO phase is linked to a stronger Aleutian low, with warmer temperatures over Alaska, northwestern Canada, and the northwestern United States, and cooler temperatures across the eastern United States (Mantua et al. 1997; Minobe 1997). Corresponding to these temperature differences are changes in precipitation, with higher amounts in the Gulf of Alaska and southwestern United States (Trenberth and Hurrell 1994; Mantua et al. 1997; Dettinger et al. 2001; Mantua and Hare 2002; Dai 2013; Mills and Walsh 2013), and geopotential heights in the troposphere favoring a deepened trough in the NP (Mantua et al. 1997; Zhang et al. 1997; Bond and Harrison 2000; Frankignoul and Sennéchael 2007). The tropospheric pattern associated with the PDO resembles the Pacific–North American (PNA) pattern, a seesaw in atmospheric pressure and geopotential heights over the NP sector that alters the circulation patterns and jet streams in the Northern Hemisphere (Wallace and Gutzler 1981). The PNA pattern consists of four centers of action: near Hawaii, over the NP, over Alberta, and over the Gulf Coast of the United States (Wallace and Gutzler 1981). The positive phase of the PNA pattern is similar to the positive phase of the PDO, with a strong Aleutian low, strong ridging over western Canada, and anomalously lower pressure over the eastern United States.
Spatial map showing the amplitude and pattern of the PDO by regressing (projecting) the monthly SST anomalies onto the monthly PC1 time series (see section 3 in the text) in the NP (°C per std dev) domain in HadISST from 1900 to 2014. The pattern shown denotes the PDO(+) phase. The amount of variance explained by the PDO is shown at the top of the figure.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
Recent research has examined the influence of extratropical SSTs and the PDO on the northern polar stratosphere. Using reanalysis data from the Modern-Era Retrospective Analysis for Research and Applications, Hurwitz et al. (2011) examined the March 2011 winter in the Northern Hemisphere and showed that the anomalously strong and cold Arctic vortex in the stratosphere could not be solely attributable to the La Niña conditions or the quasi-biennial oscillation (QBO) phase (westerly at the time). Instead, they provide evidence that warm SST anomalies in the 40°–50°N, 160°E–160°W region of the NP were responsible for the strong Arctic vortex in late winter. Following up on this work, Hurwitz et al. (2012) ran two ensembles using the Goddard Earth Observing System Chemistry–Climate Model; each ensemble consisted of 40 simulations run during an extended winter season (October–April) with specified SSTs. In each ensemble, they varied SST anomalies in the 40°–50°N, 160°E–160°W region based on winters when the central NP was warm (1990/91 and 1996/97) or cold (1986/87 and 1987/88). Climatological SSTs were used in the tropical Pacific, removing any effects from ENSO, which has been shown to weaken the polar winter vortex and lead to a greater frequency of stratospheric sudden warmings (SSWs) during El Niño (Hamilton 1993; Sassi et al. 2004; Manzini et al. 2006; Taguchi and Hartmann 2006; Camp and Tung 2007; Calvo et al. 2008; Cagnazzo et al. 2009; Free and Seidel 2009; Randel et al. 2009; Li and Lau 2013). Results showed that warm SST anomalies in the central NP lead to a strengthening of the polar vortex in late winter. They attributed this change to decreased planetary wave driving through strong ridging in the NP (Limpasuvan et al. 2005; Garfinkel et al. 2010; Nishii et al. 2010). These studies indicate that SST anomalies reminiscent of the PDO positive phase may be tied to increased planetary wave activity to induce a weaker polar vortex.
Jadin et al. (2010) provide evidence, using NCEP–NCAR reanalysis data, of a correlation of the PDO index in December with wave activity and polar vortex strength in January, suggesting a possible influence of the PDO on the stratosphere. A recent study by Woo et al. (2015) confirms a possible link between the PDO and polar vortex strength. Using NCEP–NCAR reanalysis data from 1948 to 2011, Woo et al. (2015) find a statistically significant difference in the occurrence of weak stratospheric vortex events. Weak vortex events were defined when the polar cap geopotential height anomalies at 50 hPa dropped below the 10th percentile in winter. These weak vortex events were found to occur more frequently in the PDO positive phase compared to the PDO negative phase. The increased frequency of weak vortex events is related to the tropospheric circulation anomalies in the NP, which constructively (destructively) interferes with the mean planetary wave driving during PDO positive (negative) winters.
The above studies show that there is a connection between the PDO and the polar stratosphere. Yet, no study has been able to confirm this using a long-term model simulation. While Hurwitz et al. (2012) show the influence of extratropical SST anomalies on the polar stratosphere in a model, they used climatological SSTs for the equatorial Pacific and did not consider a fully coupled ocean–atmosphere system. Furthermore, Woo et al. (2015) point out possible contamination of the PDO signal from ENSO. In this study, we investigate the influence of the PDO on the Northern Hemisphere wintertime stratosphere in a 200-yr ocean–atmosphere simulation. We account for possible contamination due to ENSO. The model we use is the National Center for Atmospheric Research (NCAR) Whole Atmosphere Community Climate Model (WACCM) (Marsh et al. 2013). The length of the simulation allows a detailed statistical analysis of the PDO’s impact on the stratosphere.
In section 2, we describe the WACCM simulation, SST dataset, and methodology. Section 3 discusses the decadal variability in NP SSTs in observations and in WACCM. Section 4 examines the influence of the PDO in WACCM on the wintertime troposphere and stratosphere. The possible contamination of the PDO signal from ENSO is also considered. Section 5 discusses the physical mechanisms involved in the stratospheric response to the PDO. Section 6 discusses the major findings from this study and their implications.
2. Methodology
a. WACCM
This study uses version 4 of the NCAR Community Earth System Model (CESM), denoted as CESM1(WACCM) (Marsh et al. 2013). For the remainder of this paper, we refer to CESM1(WACCM) as WACCM. We analyze both the atmospheric and oceanic components of this Earth system model. The atmospheric component of WACCM extends from the surface to the lower thermosphere (~140 km). It contains 66 pressure levels with a variable vertical resolution. The higher resolution of approximately 1.4 km is found in the troposphere and stratosphere; resolution in the mesosphere and thermosphere is degraded to 3.5 km. The horizontal resolution is 1.9° in latitude and 2.5° in longitude. The oceanic component of WACCM is described in detail in Deser et al. (2012). WACCM has been used to investigate the variability and frequency in SSWs (Marsh et al. 2013), effects of ENSO on the atmosphere (e.g., Marsh and Garcia 2007; Calvo et al. 2008; Randel et al. 2009), and modulation of the Holton–Tan mechanism (Holton and Tan 1980) from solar forcing and the quasi-biennial oscillation (Kren et al. 2014).
The WACCM simulation analyzed is from a 200-yr preindustrial 1850 control run. That is, greenhouse gases are fixed to 1850 preindustrial conditions, allowing us to investigate the internal and dynamical variability of the climate system without anthropogenic forcing. This is a fully coupled ocean–atmosphere run with interactive chemistry and sea ice. The simulation does not include a varying solar cycle. Solar forcing is fixed using the average solar spectral irradiance from Lean et al. (2005) over the last four solar cycles (Marsh et al. 2013). There is no quasi-biennial oscillation included in the model; instead, perpetual easterlies are present in the tropical stratosphere. Variability in volcanic stratospheric aerosols is also not included. We compare the PDO in WACCM to that analyzed by Deser et al. (2012). The main difference between the two studies is that we use a “high top” version of the Community Climate System Model, version 4 (CCSM4), that extends into the thermosphere with a lower horizontal resolution whereas Deser et al. (2012) used the low-top Community Atmospheric Model.
b. HadISST and NCEP–NCAR datasets
We first compare the WACCM PDO to the observed PDO in the Met Office Hadley Centre Sea Ice and SST (HadISST) dataset (Rayner et al. 2003); the observational analysis is limited to the period 1900–2014 because of the lack of global observations prior to 1900 (Deser et al. 2010; Messié and Chavez 2011). The HadISST dataset consists of monthly SSTs on a 1° latitude–longitude grid. After comparison of the WACCM PDO to the observed PDO, we examine the influence of the PDO on the stratosphere, including analysis of SSW frequency with PDO phase. The SSW frequency, using the method from Charlton and Polvani (2007), discussed in section 4, is shown in WACCM and compared to the observational record by including zonal-mean zonal wind at 60°N and at 10 hPa from the NCEP–NCAR reanalysis between 1948 and 2014 (Kalnay et al. 1996).
c. Analysis methods
Principal component analysis (PCA) is utilized to extract the PDO spatial pattern and frequency components in observations and in WACCM. PCA is a spectral decomposition technique to detect patterns of variability in a dataset (Wilks 2011). It has been employed in a variety of studies to examine modes of variability (e.g., Wallace and Gutzler 1981; Deser et al. 2010). Deseasonalized monthly SSTs in the region 20°–60°N, 110°E–100°W of the NP are used to extract the PDO signal. The long-term mean at each grid point is subtracted, resulting in monthly SST anomalies. For the HadISST dataset, we additionally linearly detrend the data to remove any global warming signal. This step is not performed on the WACCM simulation since a linear trend is not present because greenhouse gases in the simulation are held fixed at preindustrial levels. As in Mantua et al. (1997), empirical orthogonal functions are computed on the monthly SST anomalies over the NP and the PDO time series is defined as the leading principal component. The PDO time series is then averaged over the winter season [December–February (DJF)].
To examine the atmospheric signals that correspond to each PDO phase, we first linearly regress the WACCM DJF mean anomalies of temperature, sea level pressure (SLP), and geopotential heights onto the derived DJF mean PDO time series. Linear regression has been performed in numerous studies to examine the PDO response (e.g., Mantua et al. 1997; Zhang et al. 1997; Nigam et al. 1999; Frankignoul and Sennéchael 2007).
Composite analysis is also performed during strongly positive or negative PDO [PDO(+/−)]. PDO(+/−) winters are defined when the DJF mean PDO time series exceeds the ±1σ threshold; a similar threshold was carried out by Sassi et al. (2004) to examine the response to ENSO. Composite differences of PDO(+) minus PDO(−) are then computed and the significance of these results is determined using a Student’s t test.
As discussed in Newman et al. (2003), a large fraction of PDO variability can be explained by forcing from ENSO. Furthermore, the response of the atmosphere to the PDO is similar to the response of ENSO, with a strengthened Aleutian low, a stronger subtropical jet in the Northern Hemisphere, and a weaker polar winter vortex (e.g., Mills and Walsh 2013; Woo et al. 2015). As a result, we additionally contrast the monthly composite differences during winter in zonal-mean zonal wind and temperature of PDO(+) minus PDO(−) during ENSO neutral winters and ENSO(+) minus ENSO(−) during PDO neutral winters. A Niño-3.4 index, defined in Trenberth (1997), is calculated in WACCM over the region 5°N–5°S, 120°–170°W; ENSO(+/−) winters are defined as for the PDO when the DJF mean Niño-3.4 index exceeds the ±1σ threshold.
3. Decadal variability in North Pacific SSTs
The first principal component (PC1) is capable of representing 25% of the SST variability in HadISST and 29% in WACCM in the NP region identified as being representative of the PDO. Figure 2 shows the winter (DJF) season mean PC1 time series in WACCM and HadISST. The years in WACCM are arbitrary. Positive values above zero denote the PDO(+) phase, similar to Mantua et al. (1997), and the green line denotes a 5-yr running mean. These time series exhibit fluctuations on interannual-to-decadal time scales. However, given that the peak periodicity is at the decadal time scale (Fig. 3a), the PC1 time series are used to define the PDO, without filtering. As in Deser et al. (2012), the PC1 time series is normalized by its standard deviation. PDO(+/−) thresholds are also shown in Fig. 2. There are a total of 19 PDO(+) years and 30 PDO(−) years in WACCM, compared with 16 and 15 PDO(+/−) years in HadISST.
The normalized DJF mean PC1 time series in (a) WACCM, years 1850–1950, (b) WACCM, years 1951–2050, and (c) HadISST, years 1900–2014. The PC1 time series was derived as the normalized PC1 time series in the NP domain (20°–60°N, 110°E–100°W). The green curve denotes a 5-yr running mean, the gray solid line denotes the mean, and red and blue lines denote the thresholds (±1σ) for compositing the PDO(+) and PDO(−) winters. There are 19 PDO(+) winters and 30 PDO(−) winters in WACCM, compared with 16 PDO(+) winters and 15 PDO(−) winters in HadISST.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
(a) Power spectra of the DJF mean PC1 time series in Fig. 2 for both HadISST (red) and WACCM (blue) for the NP domain (20°–60°N, 110°E–100°W; box region). Before computing the power spectra, the time series were tapered at 10% on the ends. Also shown are spatial maps showing the amplitude and pattern of the PDO by regressing (projecting) the global DJF mean SST anomalies onto the DJF mean PC1 time series in the NP (°C per std dev) domain in both (b) HadISST and (c) WACCM. The pattern shown denotes the PDO(+) phase, with warm anomalies over the eastern and northeastern Pacific, and cold anomalies over the central–western Pacific. Percentages at the top of (b),(c) denote the percentage of variance explained by the PDO in each dataset.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
To assess the simulated PDO in WACCM, its decadal variability and spatial pattern are compared to HadISST. Figure 3a shows the power spectra of the DJF mean PDO time series in HadISST and WACCM. Comparable to past studies on the PDO variability (Messié and Chavez 2011; Deser et al. 2010, 2012), the PC1 in HadISST exhibits fluctuations on interannual-to-decadal time scales. The peak periodicity, however, occurs at decadal periods greater than 20 yr. In WACCM, this interannual-to-decadal variability is also evident, although the peak periodicity is shifted to around 15–20 yr. Figures 3b and 3c show the spatial pattern of SST anomalies associated with the PDO(+) phase in HadISST and WACCM. The spatial pattern of PDO(+) is obtained by linearly regressing the DJF mean SST anomalies onto the DJF mean PDO time series in Fig. 2. Observations show a horseshoe-shaped pattern in SST anomalies associated with the PDO, with negative SST anomalies extending from the Kuroshio–Oyashio Extension (KOE) region near Japan eastward into the central NP. The largest negative anomalies of 0.75°C are seen just south of 40°N. Positive SST anomalies are found over the eastern and northeastern NP and along the equator of the tropical Pacific Ocean; the largest anomalies of approximately 0.5°–0.75°C occur in the tropics.
In WACCM, the horseshoe-shaped SST anomaly pattern is also present, but some differences are evident. First, the amplitude in the SST anomalies over the central NP is larger than in HadISST. Negative anomalies are above 1.25°C and positive anomalies over the majority of the NP are above 0.5°C. WACCM captures the extension of these negative anomalies from the central NP westward into the KOE region. However, the largest negative anomalies are found just off the coast of Japan, as opposed to the central NP in HadISST. This difference in the PDO spatial pattern was also found by Deser et al. (2012) when analyzing the low-top version of CESM. The amplitude over the northeastern NP south of Alaska is also larger than observed. Over the tropics, the amplitude is slightly less than HadISST.
4. Winter PDO signature in the atmosphere
In the previous section, we compared features in the simulated and observed PDO. The primary focus of this paper is to examine the PDO’s influence on the polar stratosphere. Before assessing this, we first examine the response in the troposphere to show that it is similar to numerous studies using observations or other models.
a. PDO signature in the troposphere
Figure 4a shows the response in SLP during PDO(+), calculated by linearly regressing the DJF mean SLP anomalies in WACCM onto the DJF mean PDO time series shown in Fig. 2. Consistent with prior studies (e.g., Mantua et al. 1997; Zhang et al. 1997; Mantua and Hare 2002; Mills and Walsh 2013), the PDO(+) phase is marked by a stronger Aleutian low. The amplitude of the SLP change in WACCM is remarkably similar to that found in observations (Mantua et al. 1997). In WACCM, the amplitude peaks at 3.5–4 hPa; in observations, the peak is about 3 hPa. Additional negative SLP anomalies of approximately 1 hPa are found off the coast of the eastern United States, with alternate positive SLP anomalies over the Arctic region of approximately 1 hPa.
DJF mean anomalies of (a) SLP and (b) surface temperature regressed onto the DJF mean PC1 time series (Fig. 2) over the 200-yr run in WACCM. Contour intervals as identified in the respective color bars.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
Corresponding surface temperature differences are found in response to the stronger Aleutian low, shown in Fig. 4b using linear regression as for SLP. Warmer than normal surface temperatures of 0.5 K are found over the northwestern United States and up to 1 K over western Canada. The largest positive temperature anomalies, up to 2 K, are found over interior Alaska, in agreement with other studies (Mantua et al. 1997; Minobe 1997; Zhang et al. 1997; Mantua and Hare 2002; Mills and Walsh 2013). The warmer temperatures over Alaska and Canada are caused by the enhanced southerly flow from the Aleutian low. Positive anomalies between 0.25 and 0.5 K are also found over Greenland and the Arctic. Alternate cold anomalies of 0.25–0.5 K are found over far northeastern Asia, Japan, and northern Europe. Negative anomalies (0.25–0.75 K) are also found over northern Mexico, extending to the north and east into the eastern United States. Changes in precipitation (results not shown) are also found, in agreement with other studies (Power et al. 1999; Mantua et al. 1997; Mantua and Hare 2002; Wang et al. 2007, 2008; Mills and Walsh 2013).
Figure 5a shows the response to PDO(+) in the midtroposphere at 500 hPa, determined by linear regression. Negative anomalies (25 m) in geopotential heights are found in the central NP, in agreement with other studies (Mantua et al. 1997; Zhang et al. 1997; Mills and Walsh 2013), corresponding to a deeper trough connected with the stronger Aleutian low presented in Fig. 4a. Negative anomalies are also found from northeastern Mexico into the eastern United States and Atlantic Ocean. Positive height anomalies of 10–20 m are found over western Canada, Alaska, and the Arctic. The magnitude of these height differences are the same as those in observations (Mantua et al. 1997; Zhang et al. 1997), although the largest negative anomalies over the central NP seen in WACCM are found slightly south and west of those in observations. Additionally, the negative height anomalies over the United States extend farther north into southeastern Canada, whereas in observations they are confined to the far eastern United States (Mantua et al. 1997; Zhang et al. 1997). These differences, however, are not statistically different.
DJF mean geopotential height anomalies (m) at (a) 500, (b) 100, (c) 10, and (d) 3 hPa regressed onto the DJF mean PC1 time series. Contour interval as identified in the respective color bars.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
The pattern of height anomalies is reminiscent of the PNA pattern discussed earlier, with a trough over the central NP, ridging over western Canada, and a trough over the eastern United States (Wallace and Gutzler 1981). The PDO pattern resembles a wave train of anomalous geopotential heights with corresponding circulation changes and variations in the strength of the subtropical and polar jet streams. Tying Figs. 5a and 4, the ridging over western Canada and the stronger Aleutian low both act to increase surface temperatures over Alaska, Canada, the Arctic, and the northwestern United States. Another trough is found in the eastern United States, with colder temperatures. Northerly flow in this region advects cold and dry air from the north. The same response to the PDO is found when looking at the extended winter season (November–March; results not shown).
b. PDO signature in the stratosphere
In the previous section, we showed that WACCM reproduces the atmospheric response in the troposphere to the PDO. We extend our analysis in this section to the wintertime stratosphere.
Figures 5b–d show the regressions of DJF mean geopotential height anomalies at 100, 10, and 3 hPa. The height anomaly found in the troposphere at 500 hPa over Canada (Fig. 5a) is amplified at 100 hPa and reaches a magnitude of 40 m. The trough in the central NP is also amplified. Additional positive height anomalies are found over Hawaii and the subtropics of the Pacific Ocean. There are negative anomalies over the eastern United States, extending east into the Atlantic Ocean, Europe, and parts of Asia. This pattern shows that the anomaly in the troposphere is linked with a wave train of alternating height anomalies in the stratosphere. At 10 and 3 hPa, the height anomalies associated with PDO(+) are focused over the Arctic region north of 60°N. Positive height anomalies are found over northern Asia, Alaska, and northwestern Canada. Negative height anomalies are found over Greenland, eastern Canada, and northern Europe. The magnitude of the height anomalies approach 80 m at 10 hPa and 110 m at 3 hPa. The regression height pattern provides evidence that a strong anticyclone is present in the PDO(+) phase in the Arctic region, suggestive of an SSW event (Matsuno 1971).
To further examine the forcing of the PDO on the stratosphere, we perform composite analysis by grouping years into PDO(+/−) phases. Figure 6 shows the DJF mean composite differences of the 19 PDO(+) winters minus the 30 PDO(−) winters for zonal-mean zonal wind and temperature. In Fig. 6a, a stronger subtropical jet at around 30°N in the troposphere and lower stratosphere is evident in the zonal mean, with a magnitude of 5 m s−1 that is statistically significant. This is likely connected to the stronger Aleutian low in Fig. 4a. A stronger subtropical jet in the Southern Hemisphere is also present. This region of enhanced winds will be discussed in section 4c when considering effects from ENSO. At 60°N, there is a statistically significant decrease in zonal-mean zonal wind of up to 5 m s−1, implying a weaker polar jet. This decrease in the zonal-mean zonal wind extends from the troposphere into the upper stratosphere.
DJF mean composite differences of PDO(+) minus PDO(−) for (a) zonal-mean zonal wind and (b) zonal-mean temperature from 1000 to 1 hPa over all latitudes. Contour interval is 1 m s−1 and 1 K in (a) and (b), respectively. Areas stippled with black crosses denote regions that are statistically significant at or above 95% using a Student’s t test.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
In addition to zonal-mean zonal wind changes, Fig. 6b shows a warming of about 2 K over the polar stratosphere. These differences are significant at the 95% level only in the upper stratosphere around 60°N; a significant warming of 1–2 K is seen in the northern polar troposphere. Over the equatorial lower stratosphere, a statistically significant cooling of about 3 K is present, implying a strengthening of tropical upwelling and a stronger Brewer–Dobson circulation (discussed also in section 5). These zonal-mean differences are consistent with the positive phase of the PNA pattern in Baldwin and O’Sullivan (1995). The results are consistent with Woo et al. (2015) that the response of the stratosphere to the PDO(+) phase strongly resembles a wavenumber-1 stratospheric warming and is similar to the response found during the warm phase of ENSO (Sassi et al. 2004; Manzini et al. 2006; Camp and Tung 2007; Garfinkel and Hartmann 2007; Free and Seidel 2009; Cagnazzo et al. 2009).
c. Contamination between PDO and ENSO signals
When carrying out a lag correlation between the derived monthly Niño-3.4 index in WACCM and our monthly PDO time series, a correlation of 0.56 is found when ENSO leads the PDO by 8 months. This indicates that it may be difficult to distinguish between the forcings of PDO and ENSO (e.g., Newman et al. 2003; Schneider and Cornuelle 2005). The mean DJF Niño 3.4 index is 0.22 during PDO(+) and −0.69 during PDO(−) winters. There are 7 ENSO(+) winters during PDO(+) and 11 ENSO(−) winters during PDO(−). When winters that are in either the warm or cold phase of ENSO are excluded, there are 12 PDO(+) winters and 19 PDO(−) winters. Conversely, when winters that are in either the warm or cold phase of the PDO are excluded, there are 27 ENSO(+) and 37 ENSO(−) winters.
To account for possible contamination between the PDO and ENSO, we perform a seasonal composite analysis of the response of the stratosphere due to both the PDO and ENSO, when the other is in a neutral phase. Figures 7 and 8 show monthly composite differences from November through March for zonal-mean temperature and zonal wind in the stratosphere due to ENSO (during PDO neutral), PDO during ENSO neutral, and PDO during all ENSO years. For ENSO, a statistically significant warming is seen in late winter (February–March) of 4–5 K in the polar upper troposphere and lower stratosphere during ENSO(+) winters. A statistically significant difference is also seen in the tropics, with warming up to 2 K in the troposphere and 3 K colder in the tropical lower stratosphere from January through March. These changes are consistent with prior studies examining the winter response to El Niño (e.g., Calvo et al. 2008; Free and Seidel 2009). A significant decrease in the strength of the polar night jet up to 4 m s−1 is found over the middle-to-upper stratosphere during ENSO(+), maximizing in late winter (March) to 10 m s−1. In addition, the subtropical jets in the Northern and Southern Hemispheres, 20°N and 20°S in the troposphere, are strengthened. The jet in the Southern Hemisphere reaches a magnitude of 6–8 m s−1 while the northern jet reaches 12 m s−1.
(top) November–March composite differences of ENSO(+) minus ENSO(−) during PDO neutral winters for zonal-mean temperature from 1000 to 1 hPa from 30°S to 90°N. (middle) As at top, but for PDO(+) minus PDO(−) during ENSO neutral winters. (bottom) As in middle, but for PDO(+) minus PDO(−) irrespective of ENSO phase. Contour interval is 1 K; negative values are dashed and the zero contour is thickened. Yellow and orange shading denote areas that are statistically significant at 90% and 95%, respectively, using a Student’s t test.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
As in Fig. 7, but for zonal-mean zonal wind. Contour interval is 2 m s−1.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
The middle panels of Figs. 7 and 8 depict the seasonal evolution of the PDO signal during ENSO neutral, while the bottom panels show the response regardless of ENSO phase. The PDO response during ENSO neutral shows a signal that is qualitatively different from that of ENSO but has a similar magnitude. A warming of 2 K is present in early winter during PDO(+) in November; this warming becomes statistically significant in December at 60°N in the upper stratosphere. In portions of the polar troposphere, there is significant warming up to 1 K from November to December. Statistically significant cooling is present in the tropical lower stratosphere; this magnitude, however, is less than 1 K. The warming signal of the polar stratosphere of 2 K increases to 2–4 K in January from the upper troposphere to the upper stratosphere, but is no longer statistically significant. The warming signal disappears in late winter. When considering the PDO response irrespective of ENSO, similarities and differences are present. Cooling over the tropical lower stratosphere is greater than during ENSO neutral years of approximately 2 K from December through February in PDO(+) and is statistically significant. Warming of the polar troposphere and stratosphere in PDO(+) peaks in December and January as opposed to November and December during ENSO neutral. The warming is also greater at 4–5 K in January, significant at the 95% level. The warming comparably weakens in late winter as discussed earlier.
In zonal-mean zonal wind, consistent with Fig. 6, there is a statistically significant increase in the subtropical jet at 30°N during PDO(+). The magnitude of this increase is initially 2 m s−1 in November, which strengthens to 4 m s−1 in December, and decreases by late winter. Over the polar region, a statistically significant decrease in the zonal-mean zonal wind is evident of 2–4 m s−1 in the troposphere and lower stratosphere from November to December. This signal is only significant in the troposphere in January. The magnitude of the PDO response is weaker compared to when all ENSO years are included (Fig. 8, bottom). Irrespective of ENSO, the subtropical jet at 30°N is initially 2 m s−1 in November, increasing to 4–6 m s−1 from December through February. The polar jet weakening is also larger, maximizing at 4–6 m s−1, persisting into February. These zonal wind anomalies are also broader in scale compared to the PDO response during ENSO neutral years. These results support Newman et al. (2003) that a large fraction of the response to the PDO can be explained from ENSO. As a result, the zonal mean response is dominated by the ENSO signal.
The largest stratospheric response to the PDO is found in December during ENSO neutral years, as shown in the zonally varying fields in Figs. 7 and 8. Figures 9 and 10 show polar projection plots of the composite differences of PDO(+) minus PDO(−) at 500, 100, 10, and 3 hPa for zonal-mean temperature and zonal-mean zonal wind in December during ENSO neutral years. Temperatures at 500 hPa show a statistically significant warming of 2–3 K over Canada and the Arctic, with a cooling of 5 K over the central NP in connection with the deepened trough and stronger Aleutian low. At 10 and 3 hPa, the warming increases to 6 and 11 K over Eurasia and the polar region north of 60°N. This response is again very similar both in pattern and magnitude to the response to ENSO shown by Sassi et al. (2004), who found a polar warming of 13 K in February when analyzing an ensemble of three 51-yr (1950–2000) WACCM realizations using observed SSTs. The temperature response to the PDO in our study is also as large as the standard deviations found during the winter season in WACCM (not shown).
PDO(+) minus PDO(−) polar projections (during ENSO neutral years) of temperature at (a) 500, (b) 100, (c) 10, and (d) 3 hPa in December. Contour interval is 1 K at 500 and 100 hPa and 2 K at 10 and 3 hPa. Areas not stippled with white crosses denote regions that are statistically significant at or above 95% using a Student’s t test.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
As in Fig. 9, but for zonal wind. Contour interval is 2 m s−1.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
Corresponding zonal wind differences are found in Fig. 10. At 500 and 100 hPa, a stronger subtropical jet is evident in the polar projections. The maximum increase in the zonal wind (8 m s−1) is found in the central NP, tied to the location of the Aleutian low. A stronger jet over the Atlantic Ocean, as discussed in section 4a, is also present. These changes are significant at the 95% level primarily at 500 hPa. Westerlies in the NP over Alaska are reduced by 4–6 m s−1 at 500 and 100 hPa. At 10 and 3 hPa, westerlies are reduced by as much as 18 m s−1 over Eurasia. The weakened polar jet is connected with the polar warming seen in the zonal mean. In summary, while the zonal mean response to the PDO is dominated by ENSO forcing, the response during ENSO neutral in the polar projection shows a consistent picture of a warmer and weaker polar winter vortex on the regional scale.
The PDO response in the stratosphere is still present during ENSO neutral winters, yet it appears to be dominated by ENSO forcing as depicted in the zonal mean. The PDO(+) phase is conducive to a polar warming of up to 2 K in the zonal mean and a decrease of the zonal-mean zonal wind by 4 m s−1. The PDO response appears to be an early winter phenomenon, consistent with the findings of Jadin et al. (2010) that a significant response in planetary wave activity is seen in January, tied to the prior December PDO index. The ENSO response peaks in middle-to-late winter. Next, the subtropical jet strengthening seen in Fig. 6 may be tied to ENSO, whereas the increase in the northern subtropical jet appears connected to both ENSO and the PDO. Additionally, the subtropical jet increase in the Northern Hemisphere from the PDO appears at 30°–35°N, while for ENSO this increase is more equatorward at around 20°N. This may indicate a northward shift of the subtropical jet during PDO(+). Third, the magnitude of the temperature response is larger for ENSO than for the PDO. The largest regional warming over the northern pole for ENSO is 14 K in late winter March, compared to 11 K in December from the PDO. The magnitude of the PDO response is reduced during ENSO neutral. This points out that part of the PDO signal is likely tied to the uneven distribution of ENSO(+/−) events in the PDO(+/−) years, discussed further in section 6. Because of the nature of the model simulation analyzed, the present results are applicable only to the QBO easterly phase during solar average conditions. The response may be different during QBO west and/or solar maximum or minimum years. These results are consistent with Mills and Walsh (2013), who linearly remove the ENSO signal in observations and reanalysis and found that the PDO exhibits its own atmospheric signature with or without ENSO.
d. Likelihood of SSWs with PDO phase
In the previous section, we showed that the Northern Hemisphere polar stratosphere warms in early winter in the PDO(+) phase. We next examine whether an SSW is more likely to occur in either phase of the PDO and compare results to the frequency of weak stratospheric vortex events with PDO phase analyzed in Woo et al. (2015).
To calculate the number of SSWs, we use the parameters defined in Charlton and Polvani (2007). Here, major midwinter (November–March) warmings are those for which the zonal-mean zonal wind at 60°N and at 10 hPa shifts from westerly to easterly. The central date as defined by Charlton and Polvani (2007) is the day when the daily zonal-mean zonal wind at 60°N and 10 hPa first becomes easterly. To avoid double counting of SSWs, days that are within 20 days of this central date are not counted. Finally, this algorithm considers only midwinter warmings, not counting the final breakup of the polar vortex in spring. When applying this algorithm to the 200-yr WACCM simulation, we find 94 occurrences of SSWs, or about 4.7 SSWs every decade, similar to the results found by Marsh et al. (2013) when analyzing CESM. Using the PDO winters defined in section 4, 12 SSWs occur in 19 PDO(+) winters and 12 SSWs occur in 30 PDO(−) winters. Thus, 63% of the winters exhibit SSWs in the PDO(+) phase compared with 40% of the winters in the PDO(−) phase. When excluding ENSO warm and cold years, 9 SSWs occur in 12 PDO(+) winters and 10 SSWs occur in 19 PDO(−) winters. The statistics are the same, showing that now 75% of the winters exhibit SSWs in the PDO(+) phase compared to 52% in the PDO(−) phase.
These model results are compared to the observational record of the PDO from 1948 to 2014 using the NCEP–NCAR reanalysis dataset (Kalnay et al. 1996) and the observed DJF mean PDO time series from HadISST. Using the same method as for WACCM, there are a total of 37 SSWs in the 67 total winters; 7 SSWs occur in 8 PDO(+) winters and 7 in 16 PDO(−) winters. Thus, observations show a similar result, with 88% of the winters in the PDO(+) phase exhibiting SSWs compared with 44% in the PDO(−) phase. Because of the skewness and small sample size of both WACCM and NCEP–NCAR reanalysis for both phases of the PDO, this result is not statistically significant. However, this is quantitatively consistent with Woo et al. (2015), who found an increased frequency of weak stratospheric vortex events in the PDO(+) phase. Furthermore, Hurwitz et al. (2012) provide evidence for anomalously cold SSTs in the NP to alter the frequency of SSWs, agreeing with the findings in our study.
5. Physical mechanism
In this section, we investigate the physical mechanism producing this PDO response by looking at differences in the transformed Eulerian mean residual circulation and Eliassen–Palm (EP) flux (Andrews et al. 1987). We scale the EP flux vectors as in Taguchi and Hartmann (2006) by (1000/p)0.5, where p is pressure, and divide the EP flux divergence by the product of density, radius of the Earth, and cosine of latitude (Limpasuvan et al. 2004, 2005). The results shown in this section are based on composites of PDO(+/−) during ENSO neutral phases. The same conclusions apply when all years are included in the composites.
Figure 11a shows the DJF mean composite difference of PDO(+) minus PDO(−) of the zonal-mean meridional velocity υ* and vertical velocity w* between 100 and 1 hPa from the tropics to 90°N. These results show a perturbation increase in w* over the tropical stratosphere and decrease over the polar stratosphere. This indicates a strengthening of the Brewer–Dobson circulation with greater descent over the polar stratosphere. This greater sinking motion is tied to the warmer temperatures in the PDO(+) phase, as sinking motion leads to adiabatic warming.
(a) DJF mean composite differences of PDO(+) minus PDO(−) for zonal-mean residual vertical velocity (w*; contoured) between 100 and 1 hPa and 0°–90°N during ENSO neutral years. Contour interval is 0.1 mm s−1. Negative contours are dashed and zero contour is thickened. Black arrows denote the residual circulation (υ* and w*) (b) DJF mean composite differences of PDO(+) minus PDO(−) of the EP flux (arrows, momentum and heat flux components) and its divergence (contoured, contour interval is 0.5 m s−1 day−1; negative values are dashed with zero contour thickened) from 1000 to 1 hPa, also during ENSO neutral years. The EP flux vectors and divergence are scaled as in Taguchi and Hartmann (2006) and Limpasuvan et al. (2004). The EP flux and divergence is calculated from Andrews et al. (1987). Light and dark shaded areas denote regions that are statistically significant at the 95% and 99% level, respectively, using a Student’s t test.
Citation: Journal of Climate 29, 3; 10.1175/JCLI-D-15-0176.1
Figure 11b shows the composite difference, also over DJF, in the EP flux and its divergence in the zonal mean from 1000 to 1 hPa in the Northern Hemisphere. Planetary waves appear to originate from the troposphere at 45°N, the approximate location of the Aleutian low; this implies that the planetary waves originate from the NP region tied to the tropospheric circulation anomalies connected to the PDO, consistent with that found by Woo et al. (2015). These waves propagate poleward and vertically into the polar stratosphere. The strong negative EP flux divergence over the polar upper stratosphere is indicative of wave dissipation. As planetary waves dissipate, they deposit westward momentum, which acts to decelerate the zonal-mean zonal wind (Andrews et al. 1987; Limpasuvan et al. 2004; Calvo et al. 2008; Garfinkel et al. 2010; Li and Lau 2013). This coupled with a stronger mean meridional circulation leads to a polar warming event in the PDO(+) phase. These results are in agreement with earlier studies that examined varying SST anomalies in the NP and found that anomalous warm or cold SSTs in the central NP lead to changes in planetary wave activity over the polar region (Hurwitz et al. 2011, 2012). The greater planetary wave activity during PDO(+) is likely attributable to reduced atmospheric blocking from the stronger Aleutian low, which constructively interferes with planetary wave propagation (Woo et al. 2015). This mechanism is also consistent with other studies showing the influence of NP blocking patterns on wave propagation (Limpasuvan et al. 2005; Garfinkel et al. 2010; Nishii et al. 2010).
6. Summary and discussion
In this study, we have used WACCM to investigate the response of the wintertime Northern Hemisphere polar stratosphere to the PDO. We determined the DJF mean PDO time series over the NP in both WACCM and HadISST using PCA. WACCM was shown to reproduce many of the features seen in the observed PDO, including frequency variability, amplitude, and spatial pattern. The PDO(+) phase is marked by cold anomalies in the central NP and alternate warm anomalies in the northeastern and eastern NP. Variability of the PDO in observations and in WACCM shows frequency variability longer than 10 years, suggesting a role of the PDO on climate at decadal time scales.
Regressing SLP and surface temperature onto the DJF mean PDO time series showed that WACCM reproduces the observed response to the PDO. The Aleutian low strengthens in the PDO(+) phase; connected with this change is increased surface temperatures in interior Alaska, the Arctic, western Canada, and the northwestern United States. Additional differences in these variables are also found over Asia and parts of Europe, connected with the positive PNA atmospheric pattern identified in Wallace and Gutzler (1981).
The main focus of this study was to investigate the impact of the PDO on the northern pole winter stratosphere using a fully coupled 200-yr model simulation. Regressing geopotential heights in the stratosphere onto the PDO time series showed an anomalous anticyclone over western Canada and the Arctic in the PDO(+) phase, with maximum amplitude of 110 m at 3 hPa. Results showed that when DJF winters are grouped into the PDO(+/−) phases, a zonal-mean warming of up to 2 K occurs in the Northern Hemisphere polar stratosphere during the PDO(+) phase. Connected with these temperature differences is a decrease in the zonal-mean zonal wind at 60°N of about 4 m s−1.
We also investigated whether an SSW was more likely to occur in either PDO phase. Although the result is not statistically significant from the PDO composites, indications are that the PDO(+) phase exhibits a greater frequency of SSWs compared to the PDO(−) phase. This result is also seen in reanalysis data over the period 1948–2014 and is consistent with the frequency of SSWs from anomalously cold NP SSTs in Hurwitz et al. (2012) and weak stratospheric vortex events in Woo et al. (2015). The cause of this weakened polar vortex is tied to a stronger Brewer–Dobson circulation with greater descent over the poles and greater planetary wave activity from the midlatitude NP to the polar region. This finding is similar to a result found by Hurwitz et al. (2012) in which a sensitivity study using an ensemble was performed by imposing anomalous SSTs in the central NP over an extended winter season. They found a strengthening of the polar vortex when the central NP SST anomalies were anomalously warm. This would imply a strengthening of the polar vortex in the PDO(−) phase, in agreement with the opposite result found in the PDO(+) phase in our WACCM study. Our study confirms the impact of NP anomalous circulation patterns in influencing the wintertime polar vortex (Garfinkel et al. 2010; Woo et al. 2015). Furthermore, cold SST anomalies in the central NP have been shown to strengthen the Aleutian low (Frankignoul and Sennéchael 2007), which is a precursor to weakening the polar vortex (Hurwitz et al. 2012) and supporting the results found in the PDO(+) phase.
The wintertime Northern Hemisphere response in the stratosphere to the PDO exhibits the same sign as that of ENSO (e.g., Sassi et al. 2004; Taguchi and Hartmann 2006). As a result, we additionally examined the response to both the PDO and ENSO when the other was in a neutral phase. Results showed noticeable differences in the stratosphere. The magnitude of the zonal-mean warming was largest for ENSO(+) of 4–5 K in March, compared to 2 K from PDO(+). Next, the PDO(+) phase was responsible for strengthening the subtropical jet in the Northern Hemisphere, whereas the ENSO(+) phase appeared to strengthen both the northern and southern jet streams. While the magnitude of the PDO response is reduced during ENSO neutral years, the fact that a regional warming of about 11 K is still present over the polar upper stratosphere (Fig. 9) points to a decadal forcing that cannot be discounted. However, the number of years in the composites is reduced. Therefore, it is difficult to completely distinguish between ENSO and the PDO. In a future study, it would be beneficial to examine in detail the contamination between PDO and ENSO. It may be that the PDO also impacts the occurrence and magnitude of ENSO. A longer model simulation (e.g., greater than 200 yr) could be performed to improve the statistics of the results with a potentially larger sample size. Next, the PDO signal in the stratosphere was found to maximize in early winter, compared to the mid and late winter maximum response to ENSO. A future study could investigate not simply a seasonal mean (e.g., over DJF) but look at individual winter months to examine whether a strongly positive or negative PDO one month influences the polar stratosphere the subsequent month, as suggested by Jadin et al. (2010). This would aid in understanding the timing and duration of the PDO response. In summary, we suggest that the PDO signal in the stratosphere be considered a decadal ENSO-like mode. Because of the nature of the model simulation analyzed, the present results are applicable only to the QBO easterly phase during solar average conditions.
Because the PDO exhibits variability at decadal time scales, it has implications for decadal climate prediction. Since 1998, the increase in global surface temperatures has stalled, leading to the so-called “global warming hiatus” (Trenberth and Fasullo 2013; Meehl et al. 2014). Several theories have emerged to explain the leveling off of global surface temperatures over the past 15 years. One of the theories includes decadal variations in ocean temperatures manifested in the PDO (Trenberth and Fasullo 2013). It is speculated that the negative phase of the PDO after 2000 deposited more heat into the ocean below 700 m, leading to an overall warming of the oceans but a slight cooling of surface temperature (Trenberth and Fasullo 2013). Furthermore, Meehl et al. (2014) showed that while the ensemble mean of several experiments from phase 5 of the Coupled Model Intercomparison Project (CMIP5) did not accurately predict the stalling of global temperatures, several individual ensemble members accurately predicted this change when a negative PDO-like phase since 2000 was forecasted. Given this fact and what we have shown in the stratosphere, decadal climate prediction can be improved by understanding the PDO.
Studies by van Loon and Meehl (2008, 2011) examined the response of the Pacific region during years in solar cycle maxima. The sunspot number determined the maxima in the solar cycle. They found 14 solar maximum peaks from 1860 to 2000. Results showed a horseshoe-shaped pattern in SST anomalies over the NP and a weakened Aleutian low. This response, which was attributed to the solar cycle, is similar to the PDO(−) phase.
Because of the similarity of the response to the PDO shown in Fig. 4 and the response to the solar cycle found by van Loon and Meehl (2008, 2011), we examine the potential aliasing of the PDO phase with the decadal solar peaks over the observational record. To do this, we examined the HadISST DJF mean PDO time series from 1900 to 2014 shown in Fig. 2c. We also computed a January–February (JF) mean PDO time series to compare with the results of van Loon and Meehl (2008, 2011). In each solar peak year identified in van Loon and Meehl (2011) along with adding 2013, we identified whether the PDO time series was negative or positive. Table 1 shows these results. The PDO phase is predominately negative during the solar peak years. All but 2 years in the DJF mean and all but 1 year in the JF mean are in the PDO(−) phase. This finding was also confirmed by Roy and Haigh (2012) when analyzing sea level pressure data from 1856 to 2007. Thus, there is likely aliasing between the solar cycle and the PDO signals in Northern Hemisphere winter. There are several possible conclusions one could draw from this result. One possibility is that the response found by van Loon and Meehl (2008) may be due to the decadal variability in the PDO. A second possibility is that changes in the solar spectral irradiance and total solar irradiance of the sun may be driving decadal variability in the ocean (e.g., White et al. 1997; van Loon and Meehl 2011) and the resulting SST anomalies in the tropics and NP control or influence the PDO phase. Studies by Meehl et al. (2009) and Bal et al. (2011) have provided evidence that solar irradiance variations coupled with ocean–atmosphere feedbacks can impact the tropical Pacific. Changes in the tropics are also linked to the extratropics (Alexander et al. 2006). Therefore, if the PDO is influenced by solar variability, it would provide a pathway for solar cycle changes to influence climate on decadal time scales.
Solar maximum peak years as in van Loon and Meehl (2008, 2011) with 2013 added, along with the DJF and JF mean PC1 time series from the HadISST observational record (1900–2014) during the peak years. Values exceeding the ±1σ level are in boldface.
Acknowledgments
We thank the three anonymous reviewers who provided constructive comments and suggestions, which helped to greatly improve the manuscript. Part of this work was supported under NOAA Award NA09NES4400016. Resources supporting this work were provided by the NASA High-End Computing (HEC) Program through the NASA Advanced Supercomputing (NAS) Division at Ames Research Center, and NASA LWS Strategic Capability Grant NNX09AJ83G. This work also utilized the Janus supercomputer, which is supported by the National Science Foundation (Award CNS-0821794) and the University of Colorado Boulder. The Janus supercomputer is a joint effort of the University of Colorado Boulder, the University of Colorado Denver and the National Center for Atmospheric Research. NCEP Reanalysis derived data provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado, from their website at http://www.esrl.noaa.gov/psd/.
REFERENCES
Alexander, M., 2010: Extratropical air–sea interaction, sea surface temperature variability, and the Pacific decadal oscillation. Climate Dynamics: Why Does Climate Vary?, Geophys. Monogr., Vol. 189, Amer. Geophys. Union, 123–148, doi:10.1029/2008GM000794.
Alexander, M., and Coauthors, 2006: Extratropical atmosphere–ocean variability in CCSM3. J. Climate, 19, 2496–2525, doi:10.1175/JCLI3743.1.
Andrews, D. G., J. R. Holton, and C. B. Leovy, 1987: Middle Atmosphere Dynamics. Academic Press, 489 pp.
Bal, S., S. Schimanke, T. Spangehl, and U. Cubasch, 2011: On the robustness of the solar cycle signal in the Pacific region. Geophys. Res. Lett., 38, L14809, doi:10.1029/2011GL047964.
Baldwin, M. P., and D. O’Sullivan, 1995: Stratospheric effects of ENSO-related tropospheric circulation anomalies. J. Climate, 8, 649–667, doi:10.1175/1520-0442(1995)008<0649:SEOERT>2.0.CO;2.
Bond, N. A., and D. E. Harrison, 2000: The Pacific decadal oscillation, air–sea interaction and central north Pacific winter atmospheric regimes. Geophys. Res. Lett., 27, 731–734, doi:10.1029/1999GL010847.
Cagnazzo, C., and Coauthors, 2009: Northern winter stratospheric temperature and ozone responses to ENSO inferred from an ensemble of chemistry climate models. Atmos. Chem. Phys., 9, 8935–8948, doi:10.5194/acp-9-8935-2009.
Calvo, N., R. García-Herrera, and R. R. Garcia, 2008: The ENSO signal in the stratosphere. Ann. N. Y. Acad. Sci., 1146, 16–31, doi:10.1196/annals.1446.008.
Camp, C. D., and K.-K. Tung, 2007: Stratospheric polar warming by ENSO in winter: A statistical study. Geophys. Res. Lett., 34, L04809, doi:10.1029/2006GL028521.
Charlton, A. J., and L. M. Polvani, 2007: A new look at stratospheric sudden warmings. Part I: Climatology and modeling benchmarks. J. Climate, 20, 449–469, doi:10.1175/JCLI3996.1.
Chhak, K. C., E. Di Lorenzo, N. Schneider, and P. F. Cummins, 2009: Forcing of low-frequency ocean variability in the northeast Pacific. J. Climate, 22, 1255–1276, doi:10.1175/2008JCLI2639.1.
Dai, A., 2013: The influence of the inter-decadal Pacific oscillation on US precipitation during 1923–2010. Climate Dyn., 41, 633–646, doi:10.1007/s00382-012-1446-5.
Deser, C., A. S. Phillips, and J. W. Hurrell, 2004: Pacific interdecadal climate variability: Linkages between the tropics and the North Pacific during boreal winter since 1900. J. Climate, 17, 3109–3124, doi:10.1175/1520-0442(2004)017<3109:PICVLB>2.0.CO;2.
Deser, C., M. A. Alexander, S.-P. Xie, and A. S. Phillips, 2010: Sea surface temperature variability: Patterns and mechanisms. Annu. Rev. Mar. Sci., 2, 115–143, doi:10.1146/annurev-marine-120408-151453.
Deser, C., and Coauthors, 2012: ENSO and Pacific decadal variability in the Community Climate System Model version 4. J. Climate, 25, 2622–2651, doi:10.1175/JCLI-D-11-00301.1.
Dettinger, M. D., D. S. Battisti, G. J. McCabe, C. M. Bitz, and R. D. Garreaud, 2001: Interhemispheric effects of interannual and decadal ENSO-like climate variations on the Americas. Interhemispheric Climate Linkages: Present and Past Climates in the Americas and their Societal Effects, V. Markgraf, Ed., Academic Press, 1–16.
Frankignoul, C., and N. Sennéchael, 2007: Observed influence of North Pacific SST anomalies on the atmospheric circulation. J. Climate, 20, 592–606, doi:10.1175/JCLI4021.1.
Frankignoul, C., P. Muller, and E. Zorita, 1997: A simple model of the decadal response of the ocean to stochastic wind forcing. J. Phys. Oceanogr., 27, 1533–1546, doi:10.1175/1520-0485(1997)027<1533:ASMOTD>2.0.CO;2.
Free, M., and D. J. Seidel, 2009: Observed El Niño–Southern Oscillation temperature signal in the stratosphere. J. Geophys. Res., 114, D23108, doi:10.1029/2009JD012420.
Garfinkel, C. I., and D. L. Hartmann, 2007: Effects of the El Niño–Southern Oscillation and the quasi-biennial oscillation on polar temperatures in the stratosphere. J. Geophys. Res., 112, D19112, doi:10.1029/2007JD008481.
Garfinkel, C. I., D. L. Hartmann, and F. Sassi, 2010: Tropospheric precursors of anomalous Northern Hemisphere stratospheric polar vortices. J. Climate, 23, 3282–3299, doi:10.1175/2010JCLI3010.1.
Gu, D., and S. G. H. Philander, 1997: Interdecadal climate fluctuations that depend on exchanges between the tropics and extratropics. Science, 275, 805–807, doi:10.1126/science.275.5301.805.
Hamilton, K., 1993: An examination of observed Southern Oscillation effects in the Northern Hemisphere stratosphere. J. Atmos. Sci., 50, 3468–3474, doi:10.1175/1520-0469(1993)050<3468:AEOOSO>2.0.CO;2.
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, 2011: The Arctic vortex in March 2011: A dynamical perspective. Atmos. Chem. Phys., 11, 11 447–11 453, doi:10.5194/acp-11-11447-2011.
Hurwitz, M. M., P. A. Newman, and C. I. Garfinkel, 2012: On the influence of North Pacific sea surface temperature on the Arctic winter climate. J. Geophys. Res., 117, D19110, doi:10.1029/2012JD017819.
Jadin, E. A., K. Wei, Y. A. Zyulyaeva, W. Chen, and L. Wang, 2010: Stratospheric wave activity and the Pacific decadal oscillation. J. Atmos. Sol. Terr. Phys., 72, 1163–1170, doi:10.1016/j.jastp.2010.07.009.
Jin, F.-F., 1997: A theory of interdecadal climate variability of the North Pacific ocean–atmosphere system. J. Climate, 10, 1821–1835, doi:10.1175/1520-0442(1997)010<1821:ATOICV>2.0.CO;2.
Johnson, G. C., 2003: The Pacific Ocean subtropical cell surface limb. Geophys. Res. Lett., 28, 1771–1774, doi:10.1029/2000GL012723.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–470, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.
Kleeman, R., J. P. McCreary Jr., and B. A. Klinger, 1999: A mechanism for generating ENSO decadal variability. Geophys. Res. Lett., 26, 1743–1746, doi:10.1029/1999GL900352.
Kren, A. C., D. R. Marsh, A. K. Smith, and P. Pilewskie, 2014: Examining the stratospheric response to the solar cycle in a coupled WACCM simulation with an internally generated QBO. Atmos. Chem. Phys., 14, 4843–4856, doi:10.5194/acp-14-4843-2014.
Kwon, Y.-O., and C. Deser, 2007: North Pacific decadal variability in the Community Climate System Model version 2. J. Climate, 20, 2416–2433, doi:10.1175/JCLI4103.1.
Kwon, Y.-O., M. A. Alexander, N. A. Bond, C. Frankignoul, H. Nakamura, B. Qiu, and L. A. Thompson, 2010: Role of the Gulf Stream and Kuroshio–Oyashio systems in large-scale atmosphere–ocean interaction: A review. J. Climate, 23, 3249–3281, doi:10.1175/2010JCLI3343.1.
Latif, M., and T. P. Barnett, 1994: Causes of decadal climate variability over the North Pacific and North America. Science, 266, 634–637, doi:10.1126/science.266.5185.634.
Latif, M., and T. P. Barnett, 1996: Decadal climate variability over the North Pacific and North America: Dynamics and predictability. J. Climate, 9, 2407–2423, doi:10.1175/1520-0442(1996)009<2407:DCVOTN>2.0.CO;2.
Lean, J. L., G. Rottmann, J. Harder, and G. Kopp, 2005: SORCE contributions to new understanding of global change and solar variability. Sol. Phys., 230, 27–53, doi:10.1007/s11207-005-1527-2.
Li, Y., and N.-C. Lau, 2013: Influence of ENSO on stratospheric variability and the descent of stratospheric perturbations into the lower troposphere. J. Climate, 26, 4725–4748, doi:10.1175/JCLI-D-12-00581.1.
Limpasuvan, V., D. W. J. Thompson, and D. L. Hartmann, 2004: The life cycle of the Northern Hemisphere sudden stratospheric warmings. J. Climate, 17, 2584–2596, doi:10.1175/1520-0442(2004)017<2584:TLCOTN>2.0.CO;2.
Limpasuvan, V., D. L. Hartmann, D. W. J. Thompson, K. Jeev, and Y. L. Yung, 2005: Stratosphere–troposphere evolution during polar vortex intensification. J. Geophys. Res., 110, D24101, doi:10.1029/2005JD006302.
Lu, P., P. McCreary Jr., and B. A. Klinger, 1998: Meridional circulation cells and the source waters of the Pacific Equatorial Undercurrent. J. Phys. Oceanogr., 28, 62–84, doi:10.1175/1520-0485(1998)028<0062:MCCATS>2.0.CO;2.
Mantua, N. J., and S. R. Hare, 2002: The Pacific decadal oscillation. J. Oceanogr., 58, 35–44, doi:10.1023/A:1015820616384.
Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wallace, and R. C. Francis, 1997: A Pacific interdecadal climate oscillation with impacts on salmon production. Bull. Amer. Meteor. Soc., 78, 1069–1079, doi:10.1175/1520-0477(1997)078<1069:APICOW>2.0.CO;2.
Manzini, E., M. A. Giorgetta, M. Esch, L. Kornblueh, and E. Roeckner, 2006: The influence of sea surface temperatures on the northern winter stratosphere: Ensemble simulations with the MAECHAM5 model. J. Climate, 19, 3863–3881, doi:10.1175/JCLI3826.1.
Marsh, D. R., and R. R. Garcia, 2007: Attribution of decadal variability in lower-stratospheric tropical ozone. Geophys. Res. Lett., 34, L21807, doi:10.1029/2007GL030935.
Marsh, D. R., M. J. Mills, D. E. Kinnison, J.-F. Lamarque, N. Calvo, and L. M. Polvani, 2013: Climate change from 1850–2005 simulated in CESM1(WACCM). J. Climate, 26, 7372–7391, doi:10.1175/JCLI-D-12-00558.1.
Matsuno, T., 1971: A dynamical model of the stratospheric sudden warming. J. Atmos. Sci., 28, 1479–1494, doi:10.1175/1520-0469(1971)028<1479:ADMOTS>2.0.CO;2.
McPhaden, M. J., and D. Zhang, 2002: Slowdown of the meridional overturning circulation in the upper Pacific Ocean. Nature, 415, 603–608, doi:10.1038/415603a.
Meehl, G. A., J. M. Arblaster, K. Matthes, F. Sassi, and H. van Loon, 2009: Amplifying the Pacific climate system response to a small 11-year solar cycle forcing. Science, 325, 1114–1118, doi:10.1126/science.1172872.
Meehl, G. A., H. Teng, and J. M. Arblaster, 2014: Climate model simulations of the observed early-2000s hiatus of global warming. Nat. Climate Change, 4, 898–902, doi:10.1038/nclimate2357.
Messié, M., and F. Chavez, 2011: Global modes of sea surface temperature variability in relation to regional climate indices. J. Climate, 24, 4314–4331, doi:10.1175/2011JCLI3941.1.
Miller, A. J., and N. Schneider, 2000: Interdecadal climate regime dynamics in the North Pacific Ocean: Theories, observations and ecosystem impacts. Prog. Oceanogr., 47, 355–379, doi:10.1016/S0079-6611(00)00044-6.
Miller, A. J., D. R. Cayan, T. P. Barnett, N. E. Graham, and J. M. Oberhuber, 1994: Interdecadal variability of the Pacific Ocean: Model response to observed heat flux and wind stress anomalies. Climate Dyn., 9, 287–302, doi:10.1007/BF00204744.
Mills, C. M., and J. E. Walsh, 2013: Seasonal variation and spatial patterns of the atmospheric component of the Pacific decadal oscillation. J. Climate, 26, 1575–1594, doi:10.1175/JCLI-D-12-00264.1.
Minobe, S., 1997: A 50-70 year climatic oscillation over the North Pacific and North America. Geophys. Res. Lett., 24, 683–686, doi:10.1029/97GL00504.
Molinari, R. L., S. Bauer, D. Snowden, G. C. Johnson, B. Bourles, Y. Gouriou, and H. Mercier, 2003: A comparison of kinematic evidence for tropical cells in the Atlantic and Pacific Oceans. Interhemispheric Water Exchange in the Atlantic Ocean, G. J. Goni and P. Malanotte-Rizzoli, Eds., Elsevier Oceanography Series, Vol. 68, Elsevier, 269–286, doi:10.1016/S0422-9894(03)80150-2.
Newman, M., G. P. Compo, and M. A. Alexander, 2003: ENSO-forced variability of the Pacific decadal oscillation. J. Climate, 16, 3853–3857, doi:10.1175/1520-0442(2003)016<3853:EVOTPD>2.0.CO;2.
Nigam, S., M. Barlow, and E. H. Berbery, 1999: Analysis links Pacific decadal variability to drought and streamflow in United States. Eos, Trans. Amer. Geophys. Union, 80, 621–625, doi:10.1029/99EO00412.
Nishii, K., H. Nakamura, and Y. J. Orsolini, 2010: Cooling of the wintertime Arctic stratosphere induced by the western Pacific teleconnection pattern. Geophys. Res. Lett., 37, L13805, doi:10.1029/2010GL043551.
Pierce, D. W., 2001: Distinguishing coupled ocean–atmosphere interactions from background noise in the North Pacific. Prog. Oceanogr., 49, 331–352, doi:10.1016/S0079-6611(01)00029-5.
Power, S., T. Casey, C. Folland, A. Colman, and V. Mehta, 1999: Inter-decadal modulation of the impact of ENSO on Australia. Climate Dyn., 15, 319–324, doi:10.1007/s003820050284.
Randel, W. J., R. R. Garcia, N. Calvo, and D. Marsh, 2009: ENSO influence on zonal mean temperature and ozone in the tropical lower stratosphere. Geophys. Res. Lett., 36, L15822, doi:10.1029/2009GL039343.
Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108 (D14), 4407, doi:10.1029/2002JD002670.
Roy, I., and J. D. Haigh, 2012: Solar cycle signals in the Pacific and the issue of timings. J. Atmos. Sci., 69, 1446–1451, doi:10.1175/JAS-D-11-0277.1.
Sassi, F., D. Kinnison, B. A. Boville, R. R. Garcia, and R. Roble, 2004: Effect of El Niño–Southern Oscillation on the dynamical, thermal, and chemical structure of the middle atmosphere. J. Geophys. Res., 109, D17108, doi:10.1029/2003JD004434.
Schneider, N., and B. D. Cornuelle, 2005: The forcing of the Pacific decadal oscillation. J. Climate, 18, 4355–4373, doi:10.1175/JCLI3527.1.
Seager, R., Y. Kushnir, N. H. Naik, M. A. Cane, and J. Miller, 2001: Wind-driven shifts in the latitude of the Kuroshio-Oyashio Extension and generation of SST anomalies on decadal timescales. J. Climate, 14, 4249–4265, doi:10.1175/1520-0442(2001)014<4249:WDSITL>2.0.CO;2.
Shakun, J. D., and J. Shaman, 2009: Tropical origins of North and South Pacific decadal variability. Geophys. Res. Lett., 36, L19711, doi:10.1029/2009GL040313.
Solomon, A., and Coauthors, 2011: Distinguishing the roles of natural and anthropogenically forced decadal climate variability. Bull. Amer. Meteor. Soc., 92, 141–156, doi:10.1175/2010BAMS2962.1.
Taguchi, M., and D. L. Hartmann, 2006: Increased occurrence of stratospheric sudden warmings during El Niño as simulated by WACCM. J. Climate, 19, 324–332, doi:10.1175/JCLI3655.1.
Trenberth, K. E., 1997: The definition of El Niño. Bull. Amer. Meteor. Soc., 78, 2771–2777, doi:10.1175/1520-0477(1997)078<2771:TDOENO>2.0.CO;2.
Trenberth, K. E., and J. W. Hurrell, 1994: Decadal atmosphere–ocean variations in the Pacific. Climate Dyn., 9, 303–319, doi:10.1007/BF00204745.
Trenberth, K. E., and J. T. Fasullo, 2013: An apparent hiatus in global warming? Earth’s Future, 1, 19–32, doi:10.1002/2013EF000165.
van Loon, H., and G. A. Meehl, 2008: The response in the Pacific to the sun’s decadal peaks and contrasts to cold events in the Southern Oscillation. J. Atmos. Sol. Terr. Phys., 70, 1046–1055, doi:10.1016/j.jastp.2008.01.009.
van Loon, H., and G. A. Meehl, 2011: The average influence of decadal solar forcing on the atmosphere in the South Pacific region. Geophys. Res. Lett., 38, L12804, doi:10.1029/2011GL047794.
Vimont, D. J., 2005: The contribution of the interannual ENSO cycle to the spatial pattern of decadal ENSO-like variability. J. Climate, 18, 2080–2092, doi:10.1175/JCLI3365.1.
Wallace, J. M., and D. S. Gutzler, 1981: Teleconnections in the geopotential height field during the Northern Hemisphere winter. Mon. Wea. Rev., 109, 784–812, doi:10.1175/1520-0493(1981)109<0784:TITGHF>2.0.CO;2.
Wang, L., W. Chen, and R. Huang, 2007: Changes in the variability of North Pacific Oscillation around 1975/1976 and its relationship with East Asian winter climate. J. Geophys. Res., 112, D11110, doi:10.1029/2006JD008054.
Wang, L., W. Chen, and R. Huang, 2008: Interdecadal modulation of PDO on the impact of ENSO on the East Asian winter monsoon. Geophys. Res. Lett., 35, L20702, doi:10.1029/2008GL035287.
White, W. B., and D. R. Cayan, 1998: Quasi-periodicity and global symmetries in interdecadal upper ocean temperature variability. J. Geophys. Res., 103 (C10), 21 335–21 354, doi:10.1029/98JC01706.
White, W. B., J. L. Lean, D. R. Cayan, and M. D. Dettinger, 1997: Response of global upper ocean temperature to changing solar irradiance. J. Geophys. Res., 102 (C2), 3255–3266, doi:10.1029/96JC03549.
Wilks, D. S., 2011: Statistical Methods in the Atmospheric Sciences. Academic Press, 676 pp.
Woo, S.-H., M.-K. Sung, S.-W. Son, and J.-S. Kug, 2015: Connection between weak stratospheric vortex events and the Pacific decadal oscillation. Climate Dyn., 45, 3481–3492, doi:10.1007/s00382-015-2551-z.
Zhang, Y., J. M. Wallace, and D. S. Battisti, 1997: ENSO-like interdecadal variability: 1900–93. J. Climate, 10, 1004–1020, doi:10.1175/1520-0442(1997)010<1004:ELIV>2.0.CO;2.