1. Introduction
The Arctic has warmed faster than the rest of the globe during recent decades. This accelerated warming, termed Arctic amplification (AA), is strongest at the surface but is observed throughout the troposphere. Features of this warming have been linked to changes in high-latitude sea ice, snow cover, and albedo (Serreze and Francis 2006; Serreze and Barry 2011; Screen and Simmonds 2010) as well as remote influences (Screen et al. 2012; Ding et al. 2014; Sato et al. 2014). A number of studies have posited a teleconnection between Arctic warming and recent anomalous—even extreme—weather patterns in the midlatitudes (e.g., Francis and Vavrus 2012; Liu et al. 2012; Petoukhov et al. 2013; Tang et al. 2013, 2014; Coumou et al. 2014; Francis and Vavrus 2015). These patterns include the winter cold snaps in North America in 2013/14, the cold and snowy winter of 2009/10 in North America and Europe, and summer heat waves and floods on the Eurasian continent. All of these events have been linked to Arctic warming in both the scientific literature and popular media. Given the short observational record, large internal atmospheric variability, and questions surrounding the underpinning theoretical arguments, the robustness of these Arctic–midlatitude linkages is under question (Screen and Simmonds 2013; Barnes 2013; Screen et al. 2014a; Wallace et al. 2014; Barnes and Screen 2015; Hoskins and Woollings 2015). A better understanding of the underlying physical mechanisms by which Arctic warming might influence midlatitude atmospheric circulation is vital for constraining the drivers of present variability and estimates of future regional impacts of climate change.
Work investigating the potential effects of Arctic warming on the midlatitude circulation falls into two categories—one based on observational evidence (i.e., reanalyses; e.g., Francis et al. 2009; Outten and Esau 2012; Francis and Vavrus 2012; Hopsch et al. 2012; Tang et al. 2013; García-Serrano et al. 2015; King et al. 2016) and one based on climate modeling experiments (e.g., Deser et al. 2010; Kumar et al. 2010; Screen et al. 2012, 2013; Peings and Magnusdottir 2014; Deser et al. 2016). Both approaches have advanced our understanding of the problem but also point to the complexity of forcing and response in the climate system. Using observations alone, detecting robust signals is a challenge, particularly in metrics thought to be linked to extreme weather such as blocking frequency or the meridional elongation of waves (Barnes et al. 2014; Screen and Simmonds 2013). The observed changes, even if significant, may represent aliasing of other variability. Furthermore, any theory for Arctic–midlatitude linkages must account for the observed reduction in cold temperature extremes (Screen 2014) and the fact that the consequences of long-term warming are not likely to include more frequent extreme winters (Wallace et al. 2014). Using modeling experiments, many ensemble members are required for an AA-related atmospheric circulation signal to emerge owing to the large internal variability in the midlatitude atmosphere (Mori et al. 2014; Screen et al. 2014a).
Barnes and Screen (2015) propose a useful conceptual framework for discussing the Arctic’s influence on the midlatitudes, which we take inspiration from. They ask the following questions: Can it? Has it? Will it? There is consensus emerging in model-based studies that, in isolation, Arctic warming can exert an influence on midlatitude circulation (Barnes and Screen 2015), especially under projected late twenty-first century conditions (Deser et al. 2015). In the present study we make additional contributions to answering whether it can, under current conditions, and modest progress toward answering whether it has, within the limitations and scope of our idealized framework.
There is a wide range of proposed mechanisms for how the Arctic can influence the midlatitude circulation but as of yet no consensus on which are the most plausible (Cohen et al. 2014; Vihma 2014; Walsh 2014; Overland et al. 2015; Barnes and Screen 2015; Hoskins and Woollings 2015). Many studies start from the recent observed rapid sea ice decline and suggest ways the associated heating in the lower troposphere can exaggerate the waviness of the jet, which would manifest as changes in Northern Hemisphere quasi-stationary wave patterns. Sea ice loss is not the only driver of AA, or even its near-surface component, but the heating associated with it is often put forward as an important local driver of recent Arctic warming (Serreze et al. 2009; Screen and Simmonds 2010) and Arctic–midlatitude teleconnections (see, e.g., Francis and Vavrus 2012; Tang et al. 2013; Walsh 2014; Francis and Vavrus 2015). For example, the heating could weaken the equator-to-pole temperature gradient, leading to meridional stretching of the flow (e.g., Francis and Vavrus 2012), or perturb local dynamics (Rossby wave trains, cyclone paths), which can then affect the midlatitudes (e.g., Honda et al. 2009; Inoue et al. 2012). Other mechanisms invoke a deep tropospheric warming that influences the stratospheric polar vortex (Nakamura et al. 2016). Still others propose a chain of events kicked off by late-summer/autumn sea ice declines leading to Eurasian snow anomalies, upward-propagating waves, and finally a negative phase of the Arctic Oscillation (Cohen et al. 2014).
In this study we focus on the potential for recent Arctic heating in the lower troposphere to influence wintertime stationary waves. There are three reasons for this choice. 1) This pathway is regarded as one of the more plausible for Arctic–midlatitude teleconnections (Barnes and Screen 2015; Hoskins and Woollings 2015). 2) Stationary and quasi-stationary Rossby waves are the primary drivers of extratropical weather variations on monthly and longer time scales; they influence the position of the jets streams and act as guides for transient disturbances such as extratropical storms and blocking events. 3) Despite ongoing debate over how much is locally versus remotely driven, the warming in the lower troposphere is one of the most robust features of global warming to date.
Given the difficulty of establishing causality using observations or comprehensive models, a simplified approach is warranted. Idealized dynamical models are less realistic than comprehensive models but have the advantage of enabling us to test causal pathways and ascertain the robustness of physical mechanisms in ways that cannot be done with full GCMs (Held 2005). Lower-complexity models of various types have been used to investigate the atmospheric response to reduced near-surface temperature gradients (Hassanzadeh et al. 2014) and Arctic versus tropical warming (Butler et al. 2010). Using a range of models with different advantages and limitations allows us to probe different aspects of the atmospheric response to Arctic warming and build a better dynamical understanding.
In the present study we use a linear stationary wave model (SWM) with realistic topography to investigate whether wintertime midlatitude stationary waves can be modified by recent AA-related heating and the associated weakening of meridional temperature gradients in the troposphere. This type of model is well suited to address the issues around Arctic–midlatitude teleconnections as it can reproduce the large-scale atmospheric flow patterns from observations and comprehensive model simulations (Held et al. 2002) and has been used to study the response of wave patterns to both tropical and mid- to high-latitude perturbations (Ting 1996; Sobolowski et al. 2011). A particular advantage of the linear SWM employed here lies in the interpretation of cause and effect. As a result of the assumption of linearity and the limited number of forcing fields, decomposing the atmospheric response to changes in, for example, heating, is straightforward. There are limitations to such an idealized approach. One is that eddy–mean flow interactions are only included implicitly via the input fields. In other words, the model does not generate eddy–mean flow feedbacks, which may be important for any jet stream response (Hoskins and Woollings 2015). Another is that, because of the steady-state nature of the model, we are not able to capture lagged relationships such as those proposed by Cohen et al. (2014) and Honda et al. (2009). Because of these limitations, the results of the SWM must interpreted carefully and in the context of observations and more complex modeling studies. These types of investigations should be viewed as one avenue in ongoing efforts to improve our understanding of the complex dynamics at play in these types of teleconnections. However, through the simplified approach taken here we are able to investigate one of the proposed mechanisms for Arctic–midlatitude teleconnections and one direct, large-scale dynamical response to an imposed Arctic heating.
2. Methods
a. Data
The data used in this study are from the NCEP–NCAR reanalyses (Kalnay et al. 1996) at 2.5° × 2.5° horizontal resolution (17 pressure levels plus surface data). Two analysis periods are defined: a climatological period (1981–2000) and an AA period (2001–16). Both periods are within the satellite era, with the AA period covering a time of accelerated Arctic warming.
b. Model
The linear SWM used in this study is described and evaluated in detail in Ting (1991, 1994, 1996), Wang and Ting (1999), Held et al. (2002), and Sobolowski et al. (2011). A brief description is provided here.
The SWM is a baroclinic model linearized about a zonal mean basic state consisting of zonal winds u, meridional winds υ, vertical winds 
The model is idealized but retains the essential physics needed to test mechanisms relating Arctic heating to the midlatitude circulation. It is run at R30 horizontal resolution (3.75° × 2.25°) with 14 unevenly spaced sigma levels σ; at this resolution it can capture disturbances up to about wavenumber 8. It includes mass conservation and hydrostatic balance, as well as Rayleigh friction and Newtonian cooling in the boundary layer and biharmonic diffusion to eliminate small-scale noise. The forcings are considered independent, which is not precisely true in reality (Held et al. 2002), but this does allow us to isolate direct effects of the drivers of interest and assess the overall role of nonlinear interactions via the stationary nonlinear term. Explicit inclusion of the full nonlinear interactions as in the nonlinear version of the model has been shown to modulate the stationary wave patterns but not alter any of their main features (Ting et al. 2001; Sobolowski et al. 2011).
c. Experiments
Table 1 shows a summary of the experiments in this study. Though the model is run for all months we focus our analysis on the seasonal mean winter (DJF) response. The following list describes the experiments and their purpose.
Control run (CTRL): The zonal mean basic state and all forcings are from the climatological period 1981–2000.
AA period run (AAFULL): The zonal mean basic state and all forcings are from the period of enhanced Arctic warming 2001–16. This run and an associated set of decomposition experiments allow us to see which forcing terms contribute most to creating the stationary wave anomalies during the AA period.
Idealized Arctic heating experiment (AAHEAT): This experiment tests if atmospheric heating, associated with local drivers of Arctic warming, can alone directly drive a midlatitude stationary wave response. The idealized heating anomaly used to force the model is based on the wintertime temperature signature during the AA period (Figs. 1a,b). The idealized heating is applied to the Baffin Bay and Barents/Kara regions (Fig. 1c), extends up to about 700 hPa (Fig. 1d), and does not vary from month to month. The maximum heating is 1.5 K day−1 at the surface, which is consistent with observed diabatic heating anomalies, as is the prescribed depth. The basic state and three other forcing terms are taken from climatology, as in CTRL.
Atmospheric internal variability experiments (IV): These experiments are used to construct the two-standard-deviation bounds (shown in Fig. 5, described in greater detail below). The SWM is run with climatological forcings as in CTRL but with annually and monthly varying zonal mean basic states from 1980 to 2000 (20 winter seasons in total). This gives what is likely an underestimate of the envelope of internal variability but helps place the magnitude of the responses from the other experiments in context.
Sensitivity experiments were conducted to test how the response in AAHEAT depends on certain features of the idealized forcing (vertical depth, magnitude, location, and extent). For each of these experiments the heating feature of interest was varied, while the basic state and other forcing terms were taken from climatology (as in AAHEAT). There is some debate surrounding how deep the local Arctic forcing extends into the troposphere (Screen et al. 2012; Walsh 2014; Perlwitz et al. 2015; Deser et al. 2016; Nakamura et al. 2016). Therefore, the sensitivity experiments for depth cover a range from the surface to the stratosphere. For the magnitude experiments the heating anomalies were scaled such that the maximum heating at the surface was increased in 0.5-K increments up to 5 K day−1. For the location experiments the heating anomalies were shifted around the Arctic to the Beaufort Sea, East Siberian Sea, Kara Sea, Baffin Bay, and so forth. For the zonal extent experiments the heating anomalies were extended in the zonal direction (equally from their centers) from 0.5 to 3 times the extent used in AAHEAT. These experiments were then evaluated by taking the absolute value of the maximum streamfunction response for a particular sensitivity experiment between 20° and 60°N and dividing that by the absolute value of the maximum streamfunction response from CTRL between 20° and 60°N. This metric shows, as a percentage, how strong the response in a particular sensitivity experiment is relative to climatology.
Summary of experimental setup and imposed forcings. “Climatology” refers to a climatological seasonal cycle for the period 1981–2000; “AA” refers to the period 2001–16.
Wintertime (DJF) signature of the Arctic amplification period in the (a),(b) NCEP–NCAR reanalyses and (c),(d) idealized forcing used in AAHEAT. Temperature anomalies [AA period (2001–16) minus CTRL period (1981–2000); K] for (a) near-surface (σ = 0.995) and (b) vertical profile of the zonal mean. Diabatic heating anomalies representing idealized AA forcing (K day−1) for (c) near-surface (σ = 0.997) and (d) vertical profile of the zonal mean. Anomalies enclosed by solid black contours are significant at the 95% level according to a Wilcoxon rank-sum test.
Citation: Journal of Climate 29, 20; 10.1175/JCLI-D-15-0902.1
3. Results
The CTRL experiment reproduces the upper-level stationary wave patterns from reanalysis with a reasonable degree of accuracy (Figs. 2a,b). The midlatitude patterns of interest are consistent from the model to reanalysis, although there are biases in the high latitudes common to linear SWMs [see Figs. 1a,c in Held et al. (2002)]. Compared to reanalysis, the ridge over Scandinavia is weaker and contracted in the east–west direction. The North American ridge extends north into Alaska and over the Chukchi Sea in the reanalysis but not in CTRL. These deficiencies in the reproduction of climatological stationary waves in the high latitudes are generally attributed to poor scale separation at high latitudes. While not ideal, they are not too detrimental for simulating the midlatitude response to thermal forcing because the hypothesized pathway is through changes to the midlatitude winds via the thermal wind relationship. As will be seen later, the high-latitude deficiencies do not preclude Rossby waves due to Arctic perturbations from contributing to midlatitude stationary wave anomalies.
Northern Hemisphere zonally asymmetric streamfunction at σ = 0.257 (upper troposphere) for the CTRL period (1981–2000) in the (a) linear model and (b) NCEP–NCAR reanalyses. Zonally asymmetric streamfunction for the AA period (2001–16) in the (c) linear model and (d) NCEP–NCAR reanalyses. Contour interval is ±0.6 × 107 m2 s−1, with solid lines representing positive (anticyclonic flow) streamfunction values and dashed lines representing negative (cyclonic flow) streamfunction values. Thick contours indicate absolute values greater than 2.4 × 107 m2 s−1.
Citation: Journal of Climate 29, 20; 10.1175/JCLI-D-15-0902.1
To advance our understanding of which forcing terms were most important for the stationary wave responses during the AA period we conducted the AAFULL run and decomposition experiments. Figures 2c and 2d show that the SWM is able to accurately reproduce the stationary wave patterns for the AA period; Fig. 3 shows the circulation anomalies for the AA period relative to the climatological CTRL. In both the SWM simulations and reanalysis data, the AA period shows a slight strengthening of the cyclonic feature in the western subtropical Pacific and a slight weakening of both the cyclonic feature in the western subtropical Atlantic and the anticyclonic feature in the eastern Pacific off the Asian continent, with the SWM exhibiting similar biases during the AA and climatological periods (Fig. 2).1 Importantly, Fig. 3 shows that diabatic heating is the forcing responsible for the largest midlatitude response, with the stationary nonlinearity having the second-largest impact and transient forcing playing a small role. Note that the experiments shown in Fig. 3 include all factors that can influence stationary waves during the AA period, not just Arctic warming. Thus, it is no surprise that there are substantial responses in the tropics and subtropics in addition to the mid- to high latitudes (diabatic heating over the tropics/subtropics and the Canadian archipelago and nonlinear interactions over the tropics and the Greenland Sea).
Simulated response to all forcings during the AA period (2001–16) in the AAFULL experiment (color shading) compared to CTRL (contours): (a) full response, (b) response to diabatic heating, (c) response to stationary nonlinearity, and (d) response to transient heat and momentum forcing. Contour interval is ±0.3 × 107 m2 s−1.
Citation: Journal of Climate 29, 20; 10.1175/JCLI-D-15-0902.1
Idealized Arctic heating on its own (i.e., AAHEAT), produces large local responses and substantially weaker midlatitude responses (relative to the mean strength of the stationary waves and winds). A midlatitude surface pressure response between 5 and 10 hPa is evident downstream of the heating over central Europe (Fig. 4a). Within the Arctic, there is lower surface pressure (Fig. 4a) collocated with the idealized heating perturbations anomalies in AAHEAT and physically consistent anticyclonic responses in the midtroposphere (σ = 0.460) above (Fig. 4b). The AAHEAT experiment reproduces most of the high-latitude features of the anomalous circulation patterns shown in Fig. 3 but not those south of about 60°N (Fig. 4b). The midlevel response weakens the climatological trough over the Canadian archipelago and intensifies the northern flank of the climatological ridge over the North Atlantic and Europe. Elsewhere in the Arctic, away from the prescribed heating, responses of opposite sign appear, likely owing to conservation of mass. Associated with these Arctic responses are substantial changes in the winds that indicate strengthened (weakened) zonal flow over northern Greenland (the Arctic Ocean; Fig. 4c) and strengthened meridional flow over Baffin Bay and northern Siberia (Fig. 4d). In the midlatitudes, the atmospheric response to Arctic heating is modest and barotropic, which is in contrast to the baroclinic responses in the Arctic. There is a midlevel weakening of the stationary wave pattern over Europe, a modest intensification of the ridge over western North America, and a weakening of the climatological pattern extending from Africa to Asia. There is a substantial weakening of the westerly winds over the eastern North Atlantic over the jet exit region (~12 m s−1). The polar jet over the North Pacific and Alaska intensifies in response to the idealized heating (~8 m s−1). The subtropical jet is slightly weakened over the Pacific, while its northern flank over Africa and the Middle East is intensified. The only noticeable response in meridional winds is a weakening of the climatological southerlies over the Barents Sea, Scandinavia, and the British Isles.
Simulated response to idealized heating in AAHEAT (color shading) compared to climatology from CTRL (contours): (a) surface pressure (hPa), (b) σ = 0.460 zonally asymmetric streamfunction (±0.3 × 107 m2 s−1 contours; solid lines are positive/anticyclonic), (c) zonal wind at σ = 0.460 (10 m s−1 contours starting at 20 m s−1), and (d) meridional wind at σ = 0.460 (2.5 m s−1 contours; solid lines are southerlies).
Citation: Journal of Climate 29, 20; 10.1175/JCLI-D-15-0902.1
The results for AAHEAT are compared more directly to CTRL and reanalysis in Fig. 5. It is only in the high latitudes that AAHEAT (green line) exhibits an amplification of the stationary waves (Fig. 5a) relative to CTRL (blue line) beyond the two-standard-deviation bounds of internal variability during the CTRL period (gray shading, calculated from the IV experiments). In the midlatitudes and in the tropics, the AAHEAT response is within the bounds of internal variability during the CTRL period, even given that the bounds are likely an underestimate since the IV experiments only account for variability originating from year-to-year changes in the basic state (see methods). Since the SWM is steady state, we have no estimate of the variability of the AAHEAT response, but we can at least compare with the spread over the AA period from reanalysis (black dots). The high-latitude amplification in AAHEAT is actually outside the reanalysis AA period spread north of 65°N, suggesting that the pure effect of Arctic heating in AAHEAT might be mitigated by remote influences or nonlinear adjustments in the real world. In the midlatitudes, the stationary wave responses are indistinguishable, both from each other and from CTRL. Only in the tropics does the AA period response in the reanalysis deviate substantially from AAHEAT; the Arctic-only forcing in AAHEAT is clearly not able to produce the observed tropical circulation changes over the AA period. In terms of the wind responses associated with the North Pacific and North Atlantic jets (Fig. 4c), the zonal means for these two regions are shown in Figs. 5b and 5c, respectively. In the North Pacific sector, AAHEAT produces a modest weakening of the midlatitude (30°–45°N) jet within the spread of the reanalysis and a strengthening of the westerlies in high latitudes that is not evident in reanalysis, which again suggests that influences from outside the Arctic play a role in the real world. Over the North Atlantic, AAHEAT and the reanalysis exhibit similar patterns with weaker midlatitude (45°–65°N) zonal winds near the jet exit.
Comparison of wintertime zonal mean stationary wave amplitudes and zonal mean zonal winds. (a) Zonally averaged max absolute values of streamfunction at σ = 0.257 in CTRL (blue; with two-std-dev bounds from IV experiments shown in gray), AAHEAT (green), and NCEP–NCAR reanalyses for the AA period (black; dots denote individual years). Values less than (greater than) CTRL imply weaker (stronger) stationary waves. (b) Zonal mean zonal winds over the Pacific and western North America (160°E–90°W) in CTRL (blue; at σ = 0.460), AAHEAT (green; at σ = 0.460), and NCEP–NCAR reanalyses for the AA period (black; at 500 hPa). (c) As in (b), but for over the Atlantic and eastern North America (90°W–20°E).
Citation: Journal of Climate 29, 20; 10.1175/JCLI-D-15-0902.1
One possible reason that the AAHEAT response in the midlatitudes does not stand out from the variability estimated from the IV experiments is that the Arctic heating is imposed on a climatological mean basic state that is not necessarily representative of the AA period. An additional experiment, similar to AAHEAT but using the basic state for the AA period (identical to that used in AAFULL), was carried out but did not appreciably alter the results. This suggests that changes in the mean basic state between the CTRL and AA periods are not a facilitator for Arctic–midlatitude linkages. This does not rule out the possibility that basic states under future climate conditions could be more conductive to producing a midlatitude response to Arctic heating. Furthermore, within the two periods examined, individual years may have had background flow conditions that facilitated the Arctic heating’s influence on the midlatitude circulation.
Finally, the idealized nature of our Arctic heating anomaly could be questioned, despite the fact that it was designed based on observations (Fig. 1). A set of sensitivity experiments (described in the methods section) indicates that the midlatitude stationary wave response is quite sensitive to the vertical depth of Arctic heating perturbations, but not so much to their horizontal extent, strength, or location. Figure 6 shows that as the heating penetrates above the boundary layer (i.e., σ = 0.866; approximately 850 hPa), the magnitude of the response increases linearly with heating depth (Fig. 6a), although the patterns do not change (Figs. 6b–d). Increasing the heating depth from the boundary layer to the upper troposphere (σ = 0.257; approximately 250 hPa) produces a response of the same magnitude as the climatological stationary waves. Whether such a deep heating anomaly originates from local Arctic processes is the subject of some debate, as mentioned in the introduction (Screen et al. 2012; Walsh 2014; Deser et al. 2016; Nakamura et al. 2016). Recent work suggests that only ~20% of the warming in the 1000–500-hPa layer can be attributed to near-surface processes and that most of the Arctic midtropospheric warming signal originates from outside the Arctic (Perlwitz et al. 2015). Others have argued that deep heating can and does occur for specific years (Strey et al. 2010; Porter et al. 2012; Nakamura et al. 2016) or with proper representation of ocean coupling (Deser et al. 2016).
Sensitivity of wintertime (DJF) midlatitude stationary wave response at σ = 0.460 to the depth of heating anomaly: (a) max midlatitude (20°–60°N) stationary wave response as a function of depth of heating anomaly where the triangle indicates the AAHEAT experiment and the response is shown as a percentage of the climatological (i.e., CTRL) midlatitude stationary waves, (b) stationary wave response to diabatic heating anomaly extending to σ = 0.866, (c) stationary wave response to diabatic heating anomaly extending to σ = 0.568, and (d) stationary wave response to diabatic heating anomaly extending to σ = 0.257. Contours in (b)–(d) are ±0.3 × 107 m2 s−1.
Citation: Journal of Climate 29, 20; 10.1175/JCLI-D-15-0902.1
The additional sensitivity experiments for heating extent, strength, and location were performed with the heating depth held at the value used in the AAHEAT experiment (~700 hPa, indicated by the triangle in Fig. 6a). Evaluated in a manner identical to Fig. 6a, the midlatitude response to varying heating strength (Fig. 7b) and location (Fig. 7c) remains relatively flat, indicating weak sensitivity to these features of the heating perturbation. The midlatitude response is, however, somewhat sensitive to the horizontal extent of the heating (Fig. 7a) in a manner that is closely linked to the depth of the heating. That is, if the heating is shallow (i.e., <σ = 0.866) there is no sensitivity to extent; as the heating depth is increased to the level of the AAHEAT experiment and higher, the sensitivity to extent increases as well (not shown). Varying the extent and location of the heating perturbation shifts around the response pattern, and strengthening the heating perturbation intensifies the local Arctic response, but the midlatitude portion of the response only changes appreciably if the heating perturbation is deep enough. There are other measures by which the sensitivity to heating depth, extent, location, and strength could be investigated in more realistic settings, and such studies would complement these idealized experiment results.
Sensitivity of wintertime (DJF) midlatitude stationary wave response at σ = 0.460 to different aspects of the idealized AA heating anomaly: (a) horizontal extent (scaled relative to the extent of AAHEAT anomalies), (b) max heating strength (K day−1), and (c) location of heating (squares indicate anomalies shifted ~5°N). The response is defined as in Fig. 6a. Triangles indicate the AAHEAT experiment.
Citation: Journal of Climate 29, 20; 10.1175/JCLI-D-15-0902.1
4. Discussion and conclusions
In this study we explored one of the proposed physical mechanisms linking recent Arctic warming to anomalous midlatitude circulation within an idealized linear modeling framework. We focused on the wintertime stationary waves because they are so directly tied to the midlatitude jet strength and position, storms and storm tracks, and momentum and heat transport. Our major findings can be summarized as follows:
Diabatic heating is the main driver of anomalous stationary wave activity during the AA period (Fig. 3).
An idealized, but representative, diabatic heating anomaly applied over the Arctic is able to reproduce most of the high-latitude features of this response (Fig. 4).
The idealized heating alone, however, does not generate a strong midlatitude response (Figs. 4 and 5).
The overall stationary wave response is not particularly sensitive to the magnitude, location, or extent of the idealized heating but is quite sensitive to the depth of the heating, which has implications for Arctic–midlatitude teleconnections (Fig. 6).
The results generally do not support the hypothesis that recent Arctic warming has driven changes in the winter stationary waves, at least under the idealized but representative conditions prescribed here. The SWM is able to reproduce many features of the anomalous circulation throughout the Northern Hemisphere for the AA period when all the forcing terms are included (Fig. 3). However, when the forcing is limited to Arctic heating the response is also mostly limited to the Arctic. For example, wavier midlatitude jets and the necessary concomitant changes in the stationary wave patterns are not in evidence here. Simulated weakening of the jets is regionalized with the largest response at the North Atlantic jet exit region and a modest response over the Pacific. The North Atlantic weakening is seen in reanalysis but not the Pacific weakening, suggesting that other factors such as SST changes outside the Arctic may have a larger influence (Hartmann 2015). Strengthened meridional flow, which has also been linked to AA, is also limited to the high latitudes in our simulated response to Arctic heating.
These results are broadly consistent with some recent studies also aimed at separating cause and effect, by using either lower-complexity models or reanalysis. Hassanzadeh et al. (2014) used an idealized dry GCM to show that weakened meridional temperature gradients do not significantly alter wave amplitudes over a range of wavenumbers. While our experiments show some amplification of stationary wave patterns, this is confined to the high latitudes and does not extend into the midlatitudes (Fig. 5a). In future global warming scenarios, AA is shown to contribute little to projected changes in the Northern Hemisphere stationary waves (Simpson et al. 2016) and to modulate rather than dominate the midlatitude circulation response in general (Butler et al. 2010; Barnes and Polvani 2015). Finally, from reanalysis data and satellite-based flux products, Sorokina et al. (2016) show that atmospheric circulation is an important control on wintertime surface heat fluxes over the Arctic, accounting for more of the observed interannual heat flux variability than sea ice. So while changes in sea ice extent have a signature in the surface heat fluxes, the influence is not simply from the ocean to the atmosphere, suggesting that Arctic–midlatitude teleconnections involve rich, coupled interactions.
Much of the work on thermally forced Arctic–midlatitude linkages in comprehensive models consists of studies where sea ice has been perturbed (e.g., Alexander et al. 2004; Deser et al. 2010; Liu et al. 2012; Peings and Magnusdottir 2014; Screen et al. 2014a; Deser et al. 2016) both under current atmospheric conditions and future projected atmospheric conditions. These generate an atmospheric response through changes in both local thermodynamics and dynamics. The deficient stratosphere in our linear model (except insofar as it is included implicitly in the forcing fields and basic state) means that hypothesized pathways such as those suggested by some recent studies (Jaiser et al. 2013; Kim et al. 2014; Nakamura et al. 2016) will not be fully resolved in our simulations. Also the lack of nonlinear feedbacks inhibits responses that depend on interactions between eddies and the mean flow. The known shortcomings of this class of model with respect to scale separation in the high latitudes, however, appear to be less problematic for this application, as the model reproduces the high-latitude circulation anomalies reasonably well (Fig. 2). Responses that arise through subsequent dynamical adjustments or nonlinear interactions are the subject of continuing research.
Despite these limitations, Fig. 6 presents some intriguing possibilities. Indeed, it indicates that insofar as the tropospheric warming is locally induced and extends almost to the stratosphere, a midlatitude response can be generated within the linear framework employed here (although such a deep heating originating from the surface is not evident in the seasonal mean heating profile averaged over 2001–16; see Fig. 1). It is very difficult in observation-based studies to unequivocally link changes in atmospheric circulation to changes in AA-related heating (e.g., Francis and Vavrus 2012). Modeling results are not clear-cut either: typically a very large ensemble is required for a signal to emerge from the noise of internal variability, and even then the midlatitude portion of the signal is often much weaker than in reanalysis (Screen et al. 2014b; Mori et al. 2014; Blackport and Kushner 2016). Some studies identify an Arctic influence on the midlatitudes by focusing on specific years of extreme low sea ice conditions and attendant reductions in static stability, allowing surface signals to propagate through the troposphere and into the stratosphere (Strey et al. 2010; Porter et al. 2012; Kim et al. 2014). This lack of consensus points to the need for studies, such as this one, that employ simple but easily interpretable models to help separate out drivers of the dynamical complexity contained in the observational record.
Returning to the motivating questions (Can it? Has it?), the current study suggests that under specific conditions, such as those that allow deep vertical penetration of Arctic heating, the Arctic can elicit a direct response in midlatitude stationary waves in winter. Whether or not it has is a more difficult question. At least within the scope of the present study, the answer appears to be no, as our results suggest only a weak direct physical link between recent Arctic heating and midlatitude stationary waves through the proposed pathway. This does not mean that there has been no Arctic influence on midlatitude circulation patterns, only that any influence is complex, nonlinear, of uncertain magnitude, and state dependent (Trenberth et al. 2015; Overland and Wang 2015; Sorokina et al. 2016). In the long term, the externally forced signal will dominate and cold and snowy winter extremes such as those recently linked to AA will likely decrease (Screen et al. 2014b; Screen 2014; Wallace et al. 2014). In the near term, internal variability will probably still dominate, especially on regional scales. In the continuing effort to understand how climate change influences midlatitude weather patterns, it will be important then to quantify the relative influences from a range of drivers including, but not limited to, Arctic warming.
Acknowledgments
The authors wish to thank M. King, M. Ting, three anonymous reviewers, and the editor, J. Walsh, for their helpful input. This work was supported by the Nordforsk TRI project GREENICE (61841), the Research Council of Norway project jetSTREAM (231716), and the Centre for Climate Dynamics (SKD) at the Bjerknes Centre.
REFERENCES
Alexander, M. A., U. S. Bhatt, J. E. Walsh, M. S. Timlin, J. S. Miller, and J. D. Scott, 2004: The atmospheric response to realistic Arctic sea ice anomalies in an AGCM during winter. J. Climate, 17, 890–905, doi:10.1175/1520-0442(2004)017<0890:TARTRA>2.0.CO;2.
Barnes, E. A., 2013: Revisiting the evidence linking Arctic amplification to extreme weather in midlatitudes. Geophys. Res. Lett., 40, 4734–4739, doi:10.1002/grl.50880.
Barnes, E. A., and L. M. Polvani, 2015: CMIP5 projections of Arctic amplification, of the North American/North Atlantic circulation, and of their relationship. J. Climate, 28, 5254–5271, doi:10.1175/JCLI-D-14-00589.1.
Barnes, E. A., and J. A. Screen, 2015: The impact of Arctic warming on the midlatitude jet-stream: Can it? Has it? Will it? Wiley Interdiscip. Rev.: Climate Change, 6, 277–286, doi:10.1002/wcc.337.
Barnes, E. A., E. Dunn-Sigouin, G. Masato, and T. Woollings, 2014: Exploring recent trends in Northern Hemisphere blocking. Geophys. Res. Lett., 41, 638–644, doi:10.1002/2013GL058745.
Blackport, R., and P. J. Kushner, 2016: The transient and equilibrium climate response to rapid summertime sea ice loss in CCSM4. J. Climate, 29, 401–417, doi:10.1175/JCLI-D-15-0284.1.
Butler, A. H., D. W. J. Thompson, and R. Heikes, 2010: The steady-state atmospheric circulation response to climate change–like thermal forcings in a simple general circulation model. J. Climate, 23, 3474–3496, doi:10.1175/2010JCLI3228.1.
Chan, S. C., and S. Nigam, 2009: Residual diagnosis of diabatic heating from ERA-40 and NCEP reanalyses: Intercomparisons with TRMM. J. Climate, 22, 414–428, doi:10.1175/2008JCLI2417.1.
Cohen, J., and Coauthors, 2014: Recent Arctic amplification and extreme mid-latitude weather. Nat. Geosci., 7, 627–637, doi:10.1038/ngeo2234.
Coumou, D., V. Petoukhov, S. Rahmstorf, S. Petri, and H. J. Schellnhuber, 2014: Quasi-resonant circulation regimes and hemispheric synchronization of extreme weather in boreal summer. Proc. Natl. Acad. Sci. USA, 111, 12 331–12 336, doi:10.1073/pnas.1412797111.
Deser, C., R. A. Tomas, M. Alexander, and D. Lawrence, 2010: The seasonal atmospheric response to projected Arctic sea ice loss in the late twenty-first century. J. Climate, 23, 333–351, doi:10.1175/2009JCLI3053.1.
Deser, C., R. A. Tomas, and L. Sun, 2015: The role of ocean–atmosphere coupling in the zonal-mean atmospheric response to Arctic sea ice loss. J. Climate, 28, 2168–2186, doi:10.1175/JCLI-D-14-00325.1.
Deser, C., L. Sun, R. A. Tomas, and J. Screen, 2016: Does ocean coupling matter for the northern extratropical response to projected Arctic sea ice loss? Geophys. Res. Lett., 43, 2149–2157, doi:10.1002/2016GL067792.
Ding, Q., J. M. Wallace, D. S. Battisti, E. J. Steig, A. J. E. Gallant, H.-J. Kim, and L. Geng, 2014: Tropical forcing of the recent rapid Arctic warming in northeastern Canada and Greenland. Nature, 509, 209–212, doi:10.1038/nature13260.
Francis, J. A., and S. J. Vavrus, 2012: Evidence linking Arctic amplification to extreme weather in mid-latitudes. Geophys. Res. Lett., 39, L06801, doi:10.1029/2012GL051000.
Francis, J. A., and S. J. Vavrus, 2015: Evidence for a wavier jet stream in response to rapid Arctic warming. Environ. Res. Lett., 10, 014005, doi:10.1088/1748-9326/10/1/014005.
Francis, J. A., W. Chan, D. J. Leathers, J. R. Miller, and D. E. Veron, 2009: Winter Northern Hemisphere weather patterns remember summer Arctic sea-ice extent. Geophys. Res. Lett., 36, L07503, doi:10.1029/2009GL037274.
García-Serrano, J., V. Guemas, and F. Doblas-Reyes, 2015: Added-value from initialization in predictions of Atlantic multi-decadal variability. Climate Dyn., 44, 2539–2555, doi:10.1007/s00382-014-2370-7.
Hartmann, D. L., 2015: Pacific sea surface temperature and the winter of 2014. Geophys. Res. Lett., 42, 1894–1902, doi:10.1002/2015GL063083.
Hassanzadeh, P., Z. Kuang, and B. F. Farrell, 2014: Responses of midlatitude blocks and wave amplitude to changes in the meridional temperature gradient in an idealized dry GCM. Geophys. Res. Lett., 41, 5223–5232, doi:10.1002/2014GL060764.
Held, I. M., 2005: The gap between simulation and understanding in climate modeling. Bull. Amer. Meteor. Soc., 86, 1609–1614, doi:10.1175/BAMS-86-11-1609.
Held, I. M., M. Ting, and H. Wang, 2002: Northern winter stationary waves: Theory and modeling. J. Climate, 15, 2125–2144, doi:10.1175/1520-0442(2002)015<2125:NWSWTA>2.0.CO;2.
Honda, M., J. Inoue, and S. Yamane, 2009: Influence of low Arctic sea-ice minima on anomalously cold Eurasian winters. Geophys. Res. Lett., 36, L08707, doi:10.1029/2008GL037079.
Hopsch, S., J. Cohen, and K. Dethloff, 2012: Analysis of a link between fall Arctic sea ice concentration and atmospheric patterns in the following winter. Tellus, 64, 18624, doi:10.3402/tellusa.v64i0.18624.
Hoskins, B., and T. Woollings, 2015: Persistent extratropical regimes and climate extremes. Current Climate Change Rep., 1, 115–124, doi:10.1007/s40641-015-0020-8.
Inoue, J., M. E. Hori, and K. Takaya, 2012: The role of Barents Sea ice in the wintertime cyclone track and emergence of a warm-Arctic cold-Siberian anomaly. J. Climate, 25, 2561–2568, doi:10.1175/JCLI-D-11-00449.1.
Jaiser, R., K. Dethloff, and D. Handorf, 2013: Stratospheric response to Arctic sea ice retreat and associated planetary wave propagation changes. Tellus, 65A, 19375, doi:10.3402/tellusa.v65i0.19375.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.
Kim, B.-M., S.-W. Son, S.-K. Min, J.-H. Jeong, S.-J. Kim, X. Zhang, T. Shim, and J.-H. Yoon, 2014: Weakening of the stratospheric polar vortex by Arctic sea-ice loss. Nat. Commun., 5, 4646, doi:10.1038/ncomms5646.
King, M., M. Hell, and N. Keenlyside, 2016: Investigation of the atmospheric mechanisms related to the autumn sea ice and winter circulation link in the Northern Hemisphere. Climate Dyn., 46, 1185–1195, doi:10.1007/s00382-015-2639-5.
Kumar, A., and Coauthors, 2010: Contribution of sea ice loss to Arctic amplification. Geophys. Res. Lett., 37, L21701, doi:10.1029/2010GL045022.
Ling, J., and C. Zhang, 2013: Diabatic heating profiles in recent global reanalyses. J. Climate, 26, 3307–3325, doi:10.1175/JCLI-D-12-00384.1.
Liu, J., J. A. Curry, H. Wang, M. Song, and R. M. Horton, 2012: Impact of declining Arctic sea ice on winter snowfall. Proc. Natl. Acad. Sci. USA, 109, 4074–4079, doi:10.1073/pnas.1114910109.
Mori, M., M. Watanabe, H. Shiogama, J. Inoue, and M. Kimoto, 2014: Robust Arctic sea-ice influence on the frequent Eurasian cold winters in past decades. Nat. Geosci., 7, 869–873, doi:10.1038/ngeo2277.
Nakamura, T., K. Yamazaki, K. Iwamoto, M. Honda, Y. Miyoshi, Y. Ogawa, Y. Tomikawa, and J. Ukita, 2016: The stratospheric pathway for Arctic impacts on midlatitude climate. Geophys. Res. Lett., 43, 3494–3501, doi:10.1002/2016GL068330.
Outten, S. D., and I. Esau, 2012: A link between Arctic sea ice and recent cooling trends over Eurasia. Climatic Change, 110, 1069–1075, doi:10.1007/s10584-011-0334-z.
Overland, J., and M. Wang, 2015: Increased variability in the early winter subarctic North American atmospheric circulation. J. Climate, 28, 7297–7305, doi:10.1175/JCLI-D-15-0395.1.
Overland, J., J. A. Francis, R. Hall, E. Hanna, S.-J. Kim, and T. Vihma, 2015: The melting Arctic and midlatitude weather patterns: Are they connected? J. Climate, 28, 7917–7932, doi:10.1175/JCLI-D-14-00822.1.
Peings, Y., and G. Magnusdottir, 2014: Response of the wintertime Northern Hemisphere atmospheric circulation to current and projected Arctic sea ice decline: A numerical study with CAM5. J. Climate, 27, 244–264, doi:10.1175/JCLI-D-13-00272.1.
Perlwitz, J., M. Hoerling, and R. Dole, 2015: Arctic tropospheric warming: Causes and linkages to lower latitudes. J. Climate, 28, 2154–2167, doi:10.1175/JCLI-D-14-00095.1.
Petoukhov, V., S. Rahmstorf, S. Petri, and H. J. Schellnhuber, 2013: Quasiresonant amplification of planetary waves and recent Northern Hemisphere weather extremes. Proc. Natl. Acad. Sci. USA, 110, 5336–5341, doi:10.1073/pnas.1222000110.
Porter, D. F., J. J. Cassano, and M. C. Serreze, 2012: Local and large-scale atmospheric responses to reduced Arctic sea ice and ocean warming in the WRF Model. J. Geophys. Res., 117, D11115, doi:10.1029/2011JD016969.
Sato, K., J. Inoue, and M. Watanabe, 2014: Influence of the Gulf Stream on the Barents Sea ice retreat and Eurasian coldness during early winter. Environ. Res. Lett., 9, 084009, doi:10.1088/1748-9326/9/8/084009.
Screen, J. A., 2014: Arctic amplification decreases temperature variance in northern mid- to high-latitudes. Nat. Climate Change, 4, 577–582, doi:10.1038/nclimate2268.
Screen, J. A., and I. Simmonds, 2010: The central role of diminishing sea ice in recent Arctic temperature amplification. Nature, 464, 1334–1337, doi:10.1038/nature09051.
Screen, J. A., and I. Simmonds, 2013: Exploring links between Arctic amplification and mid-latitude weather. Geophys. Res. Lett., 40, 959–964, doi:10.1002/grl.50174.
Screen, J. A., C. Deser, and I. Simmonds, 2012: Local and remote controls on observed Arctic warming. Geophys. Res. Lett., 39, L10709, doi:10.1029/2012GL051598.
Screen, J. A., I. Simmonds, C. Deser, and R. Tomas, 2013: The atmospheric response to three decades of observed Arctic sea ice loss. J. Climate, 26, 1230–1248, doi:10.1175/JCLI-D-12-00063.1.
Screen, J. A., C. Deser, I. Simmonds, and R. Tomas, 2014a: Atmospheric impacts of Arctic sea-ice loss, 1979–2009: Separating forced change from atmospheric internal variability. Climate Dyn., 43, 333–344, doi:10.1007/s00382-013-1830-9.
Screen, J. A., C. Deser, and L. Sun, 2014b: Reduced risk of North American cold extremes due to continued Arctic sea ice loss. Bull. Amer. Meteor. Soc., 96, 1489–1503, doi:10.1175/BAMS-D-14-00185.1.
Serreze, M. C., and J. Francis, 2006: The Arctic amplification debate. Climatic Change, 76, 241–264, doi:10.1007/s10584-005-9017-y.
Serreze, M. C., and R. G. Barry, 2011: Processes and impacts of Arctic amplification: A research synthesis. Global Planet. Change, 77, 85–96, doi:10.1016/j.gloplacha.2011.03.004.
Serreze, M. C., A. Barrett, J. Stroeve, D. Kindig, and M. Holland, 2009: The emergence of surface-based Arctic amplification. Cryosphere, 3, 11–19, doi:10.5194/tc-3-11-2009.
Simpson, I. R., R. Seager, M. Ting, and T. A. Shaw, 2016: Causes of change in Northern Hemisphere winter meridional winds and regional hydroclimate. Nat. Climate Change, 6, 65–70, doi:10.1038/nclimate2783.
Sobolowski, S., G. Gong, and M. Ting, 2011: Investigating the linear and nonlinear stationary wave response to anomalous North American snow cover. J. Atmos. Sci., 68, 904–917, doi:10.1175/2010JAS3581.1.
Sorokina, S. A., C. Li, J. J. Wettstein, and N. G. Kvamstø, 2016: Observed atmospheric coupling between Barents Sea ice and the warm-Arctic cold-Siberian anomaly pattern. J. Climate, 29, 495–511, doi:10.1175/JCLI-D-15-0046.1.
Strey, S. T., W. L. Chapman, and J. E. Walsh, 2010: The 2007 sea ice minimum: Impacts on the Northern Hemisphere atmosphere in late autumn and early winter. J. Geophys. Res., 115, D23103, doi:10.1029/2009JD013294.
Tang, Q., X. Zhang, X. Yang, and J. A. Francis, 2013: Cold winter extremes in northern continents linked to Arctic sea ice loss. Environ. Res. Lett., 8, 014036, doi:10.1088/1748-9326/8/1/014036.
Tang, Q., X. Zhang, and J. A. Francis, 2014: Extreme summer weather in northern mid-latitudes linked to a vanishing cryosphere. Nat. Climate Change, 4, 45–50, doi:10.1038/nclimate2065.
Ting, M., 1991: The stationary wave response to a midlatitude SST anomaly in an idealized GCM. J. Atmos. Sci., 48, 1249–1275, doi:10.1175/1520-0469(1991)048<1249:TSWRTA>2.0.CO;2.
Ting, M., 1994: Maintenance of northern summer stationary waves in a GCM. J. Atmos. Sci., 51, 3286–3308, doi:10.1175/1520-0469(1994)051<3286:MONSSW>2.0.CO;2.
Ting, M., 1996: Steady linear response to tropical heating in barotropic and baroclinic models. J. Atmos. Sci., 53, 1698–1709, doi:10.1175/1520-0469(1996)053<1698:SLRTTH>2.0.CO;2.
Ting, M., H. Wang, and L. Yu, 2001: Nonlinear stationary wave maintenance and seasonal cycle in the GFDL R30 GCM. J. Atmos. Sci., 58, 2331–2354, doi:10.1175/1520-0469(2001)058<2331:NSWMAS>2.0.CO;2.
Trenberth, K. E., J. T. Fasullo, and T. G. Shepherd, 2015: Attribution of climate extreme events. Nat. Climate Change, 5, 725–730, doi:10.1038/nclimate2657.
Vihma, T., 2014: Effects of Arctic sea ice decline on weather and climate: A review. Surv. Geophys., 35, 1175–1214, doi:10.1007/s10712-014-9284-0.
Wallace, J. M., I. M. Held, D. W. Thompson, K. E. Trenberth, and J. E. Walsh, 2014: Global warming and winter weather. Science, 343, 729–730, doi:10.1126/science.343.6172.729.
Walsh, J. E., 2014: Intensified warming of the Arctic: Causes and impacts on middle latitudes. Global Planet. Change, 117, 52–63, doi:10.1016/j.gloplacha.2014.03.003.
Wang, H., and M. Ting, 1999: Seasonal cycle of the climatological stationary waves in the NCEP–NCAR reanalysis. J. Atmos. Sci., 56, 3892–3919, doi:10.1175/1520-0469(1999)056<3892:SCOTCS>2.0.CO;2.
Wang, L., and P. J. Kushner, 2011: Diagnosing the stratosphere–troposphere stationary wave response to climate change in a general circulation model. J. Geophys. Res., 116, D16113, doi:10.1029/2010JD015473.
AAFULL does not exhibit the ridge–trough anomaly pattern over the Barents/Kara Sea and East Asia that has been a focus of many recent studies. This pattern emerges from composites of individual winters based on a detrended sea ice index (e.g., Honda et al. 2009; Inoue et al. 2012), not as a time-mean feature of the AA period, which is what the experiment was designed to capture. The ridge–trough anomaly is reproduced by the model in experiments isolating the thermal forcing related to Barents Sea ice loss for individual winters (not shown).