Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The influence of Arctic amplification on mid-latitude summer circulation

The influence of Arctic amplification on mid-latitude summer circulation REVIEW ARTICLE DOI: 10.1038/s41467-018-05256-8 OPEN The influence of Arctic amplification on mid-latitude summer circulation 1,2 1,2 3 4 5 D. Coumou , G. Di Capua , S. Vavrus , L. Wang & S. Wang Accelerated warming in the Arctic, as compared to the rest of the globe, might have profound impacts on mid-latitude weather. Most studies analyzing Arctic links to mid-latitude weather focused on winter, yet recent summers have seen strong reductions in sea-ice extent and snow cover, a weakened equator-to-pole thermal gradient and associated weakening of the mid-latitude circulation. We review the scientific evidence behind three leading hypotheses on the influence of Arctic changes on mid-latitude summer weather: Weakened storm tracks, shifted jet streams, and amplified quasi-stationary waves. We show that interactions between Arctic teleconnections and other remote and regional feedback processes could lead to more persistent hot-dry extremes in the mid-latitudes. The exact nature of these non- linear interactions is not well quantified but they provide potential high-impact risks for society. he observed increases in the frequency and intensity of extreme heat and heavy rainfall events since the late 1980s, especially in mid-latitude regions, have been linked to 1–3 Tanthropogenic global warming . Scientists are generally confident in the thermo- 4,5 dynamic drivers of these changes but are less so in dynamic aspects . Another pronounced signal of anthropogenic global warming is the rapidly increasing near-surface temperatures in the Arctic at a pace two to four times faster than the rest of the globe, known as Arctic amplification (AA) . The extent to which AA affects the mid-latitude circulation and possibly contribute to the observed increases in weather extremes has been a subject of active debate . Most studies analyzing the role of AA on mid-latitude weather have focused on the winter season and the linkage with cold spells. The stronger jet stream, the presence of the stratospheric polar vortex, and the post-1990s increase in abnormally cold winters over central Eurasia have 6,8,9 drawn a lot of attention to the winter season . The increased heat stored in the Arctic Ocean owing to sea-ice loss is released into the atmosphere in early winter. The associated expansion of the near-surface air increases Arctic geopotential heights and can affect the circumglobal cir- culation directly as well as via feedbacks between the troposphere and stratosphere involving the 6,10–14 stratospheric polar vortex . Even though the exact pathways through which the Arctic influences the mid-latitude winter circulation are debated, a scientific consensus is emerging that 7,15,16 AA has at least some influence on winter weather . Links between AA and summer circulation have received far less scientific attention, despite the potential for synergistic effects that might favor high-impact extremes. In summer, thermodynamic 1 2 Department of Water & Climate Risk, Institute for Environmental Studies, VU Amsterdam, Amsterdam 1087HV, Netherlands. Department of Earth System Analyses, Potsdam Institute for Climate Impact Research, Potsdam 14473, Germany. Nelson Institute Center for Climatic Research, University of Wisconsin-Madison, Madison 53706 WI, USA. Department of Earth and Planetary Sciences, Harvard University, Cambridge 02138 MA, USA. Department of Plants, Soils and Climate, Utah State University, Logan 84322 UT, USA. Correspondence and requests for materials should be addressed to D.C. (email: coumou@pik-potsdam.de) NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 1 1234567890():,; REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 BOX 1 | Recent mid-latitude summer weather extremes and their impacts Many recent high-impact summer heatwaves occurred in the far-tail of the distribution and are difficult to explain by the direct radiative warming effect 18–21 of greenhouse gas forcing alone . In 2010, Russia saw 33 consecutive hot-and-dry days (with temperatures above 30 °C), resulting in an estimated 55,000 heat-related deaths, more than 500 wildfires near Moscow and grain-harvest losses of 30% . A quantitative global analyses showed that the 2010 event was the most-severe heatwave ever recorded worldwide, based on a heatwave index that can be used across different regions . Intriguingly, all record-setting heatwaves based on this index occurred in the mid-latitudes, indicating that here heatwaves are becoming more intense at a pace that exceeds the global mean . Extreme summer heat in the Northern Hemisphere mid-latitudes now far exceeds historical frequencies in the twentieth century . Over the last decade, Europe has seen an exceptionally rapid increase in the chance of extremely hot summers similar to the 2003 21 4,20 extreme . Other notable high-impact and record-breaking droughts and heatwaves occurred in the USA in 2011 and 2012 , leading to billions of 120,121 dollars in agricultural losses . One reason why these extremes cause so much damage is that temperature and precipitation during summer are anti-correlated virtually everywhere across extratropical land . Consequently, extremes of heat and dryness often coincide to produce compound extremes that exert disproportionately 123 124 large societal impacts . Certain mid-latitude regions identified as hotspots of tight atmosphere–land coupling, such as central North America , are especially prone to concurrent heatwaves and droughts. Such events are promoted by positive thermodynamic feedbacks that favor depleted soil moisture and enhanced sensible surface heating as the land warms, but they are also strongly regulated by atmospheric dynamics that initiate and sustain anomalous heating and drying . These kinds of summertime compound events are therefore highly relevant for the dynamical changes described in this study, and recent studies have developed 126–128 innovative techniques to separate the contributions from dynamics and thermodynamics . In particular, the projected trend toward a weaker and poleward-shifted jet stream is consistent with projections of a significantly increased risk of compound hot-dry extremes across much of the Northern Hemisphere this century . This type of climate change would likely exacerbate the separate impacts of extreme heat and dryness, based on the 129 130 131 documented stresses that compound heatwaves and droughts exert in causing disease , vegetation mortality , wildfires , and agricultural losses . and dynamic drivers of extreme weather could act in the same when near-surface air temperatures drop below sea-surface direction, leading to tail risks . For instance, any increased fre- temperatures, this excessive heat is released into the 6,10–13 quency in circulation regimes conducive to persistent heat extremes atmosphere . The additional heat inflates the lower tropo- would act on top of the thermodynamically driven increase in heat, sphere over the Arctic Ocean and nearby continents, and increase creating possibilities for very-extreme heatwaves. Many recent high- geopotential heights, which could affect circulation patterns fur- impact summer heatwaves indeed occurred in that far-tail of the ther south. 6,14 distribution and cannot be explained by the direct thermodynamic The observed increase in Arctic geopotential heights might 18–21 effect of greenhouse gas forcing alone (Box 1) .Suchextreme reduce the poleward pressure gradient in the troposphere and heatwaves have been found to increase and intensify across most therefore weaken the storm tracks and westerly jet. However, this regions but more so in the mid-latitudes than over the rest of the notion based on thermal-wind balance and baroclinicity provides globe . Consistent with the increase in heatwaves, the hot tail of little explanation for the recent changes in winter circulation. In summer temperature distribution has been warming faster than the winter, the near-surface warming in the Arctic has been pro- median and the cold tail. Figure 1 shows the warming trends in the nounced but confined to high latitudes only, i.e., north of 70° N . 95th percentile (hot tail), 50th percentile (median), and 5th per- Within the mid-latitudes (i.e., 30° N–60° N), neither the poleward centile (cold tail) of daily summer temperatures. Clearly, over most temperature gradient nor the zonal-mean jet or storm track have mid-latitude regions, in particular over Eurasia but less so in the US, seen any significant changes in winter . Future climate model the hot tail has been warming faster than the cold tail and thus projections under high-emission scenarios show that both chan- 22 27,28 29–31 temperature variability in summer has increased . This increased ges in the tropics and in the Arctic can influence the variability indicates that more complex processes beyond simple strength and position of the mid-latitude winter circulation. radiative greenhouse gas forcing are important in driving heat Enhanced warming projected in the tropical tropopause region extremes (Box 1). This is supported by recent studies that indicate (due to enhanced deep convection and latent heating) acts to that summer weather has become more persistent in several regions increase the upper-level poleward temperature gradient, which 23–25 28 in the mid-latitudes . In summer, the hot tail of the distribution strengthens the mid-latitude westerlies . This has become known is associated with persistent, blocking weather systems, and an as the tug-of-war (Fig. 2), whereby tropical changes tend to increase in their persistence leads to more extreme temperatures. strengthen mid-latitude circulation and lead to a poleward Here, we review recent studies analyzing possible links between migration, whereas AA has the opposite effect . AA, mid-latitude summer weather and extreme events, in particular Many recent winters were characterized by extremely warm persistent hot-dry extremes. We start by giving a brief synopsis of temperatures in the Arctic and anomalously cold conditions the different Arctic mechanisms proposed for winter. Next, we further south, especially over Eurasia. In fact, large areas over address the seasonal differences in the mid-latitude circulations, in central Eurasia have been cooling since 1990 . Possible dyna- the influence of regional and far-away drivers, and in detected mical mechanisms behind this warm-Arctic cold-continent changes in the Arctic. We focus on three possible dynamical pattern involving sea-ice loss consist of a direct tropospheric pathways that are most relevant to summer and summarize the pathway and pathways involving the stratospheric polar theoretical, empirical, and modeling evidence for each of them. We vortex (Fig. 2) . In the latter hypothesis, increased geopotential discuss the confidence and uncertainties associated with these heights over high-latitude regions can cause a pronounced dynamical pathways, identify knowledge gaps and key societal risks, upward wave propagation into the stratosphere which can and provide a roadmap for future research. weaken the stratospheric polar vortex and, in extreme cases, trigger sudden stratospheric warming events. A weak polar vortex can propagate downwards into the troposphere causing a Arctic amplification and mid-latitude winter circulation negative Arctic oscillation (AO) that is conducive to cold spells Due to declining sea-ice, the Arctic Ocean absorbs more in Eurasia and Siberia . incoming solar radiation from spring to autumn. By early winter, 2 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications U250 hPa U250 hPa v250 hPa v250 hPa T2m T2m U30 hPa NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE a d 95% 95 – 50% qqq q q q q q q q q q q q q q q q qq q q q q q q q q q qq q q q q qq q q q q q q q q q qq q q q q q q q q q q q qq q q q q q q q q qq q q q qq q q q q q q q q qq q q q qq qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q qqq q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q b e 50% 95 – 5% q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qqq q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q qq q qq q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q qq q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q c f 5% 50 – 5% q q qqq q qqq q q q q q q q q q q q qq q q q q q q q q q q q q q q q qq q q q qq q q q q q q q q qq q q q q q qq q q q q q q q qq q q q qq q q q q qq q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q −1.5 −1.2 −0.9 −0.6 −0.3 0 0.3 0.6 0.9 1.2 1.5 °C/decade Fig. 1 Summer trends in surface temperature over 1980–2011. a 95th, b 50th, and c 5th quantile of the HadGHCND gridded daily dataset; differences in the trends of different quantiles, plotted in d–f,reflect changes in the width of the distribution. Over most mid-latitude regions, especially over Eurasia, the width of the distribution has broadened and thus variability has increased. (Figure created using R statistical software) Winter Summer ab Monthly long term mean wind speed (m/s) January 20.0 30.0 40.0 50.0 60.0 70.0 July 12.0 16.0 20.0 24.0 28.0 32.0 Stronger jets & a polar vortex Weaker subtropical jet & double jets cd Monthly long term mean v wind (m/s) January –12.5 –7.5 –2.5 2.5 7.5 12.5 July –7.5 –4.5 –1.5 1.5 4.5 7.5 January T2m 230.0 244.0 258.0 272.0 286.0 300.0 July T2m 265.0 272.0 279.0 286.0 293.0 300.0 Longer stationary waves Shorter stationary waves Weaker land–sea heat contrast Stronger land–sea heat contrast Fig. 2 Schematic figure illustrating the main seasonal differences in upper tropospheric circulation between winter (January) and summer (July). Panels a and b show 250-hPa wind speed (green-to-blue shading) illustrating the jet streams with black arrow lines that follow the zone of maximum wind speed. The wintertime stratospheric polar vortex is outlined with the thick green line following the 30-hPa maximum wind speed. Panels c and d show the 250 hPa meridional wind speed (dark gray-to-dark red shading) depicting the stationary wave features associated with the jet streams. White arrows are addedto illustrate wind direction. Basic differences in the summer circulation features, as compared to winter, include shorter stationary waves, more northerly subtropical jet, absence of stratospheric polar vortex and an Arctic front jet forming double jets. Data are 1970–2000 climatology of NCEP Reanalysis (downloadable: https://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html). (Figure created using Panoply and Apple’s Keynote software) NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 3 REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 There is relatively high confidence that this stratospheric and severe weather outbreaks (e.g., flooding) due to their 56,57 pathway is a real phenomenon and has a role in the warm- longer lifetime than synoptic disturbances . Arctic cold-continent pattern, as it is supported by The climatological jets in summer have less northward tilt multiple lines of evidence: From empirical analyses, to causal compared to winter and therefore waveguides are also oriented 6,15,35,36 discovery algorithms, and climate model simulations . west-to-east and have the potential to become circumglobal Still, its relative importance compared to other pathways, (Fig. 2). Also, the narrower mean jet in summer favors the tropical influences and atmosphere internal variability remains waveguide effect with wave trains orienting zonally along the jet 7 38,56,58,59 unclear . stream waveguide . However, the zonal orientation of the climatological jet does not mean that the overall flow is less wavy in summer compared to winter: The total waviness in geopo- Summer circulation tential height fields on sub-synoptic to sub-seasonal time scales is 60–62 Compared to winter, summer circulation in the mid-latitude is as pronounced in summer as in winter, if not more .In weaker, more barotropic and the climatological jets are more winter, Rossby waves are typically oriented along a meridional zonally oriented, which promotes the formation of circumglobal path arcing from the tropics into the mid-latitudes, or from the wave trains (CGWT, see Box 2). It is less influenced by variability mid-latitudes into the tropics (see Fig. 3 of Hoskins and Wool- in tropical sea-surface temperatures (SST), and more sensitive to lings ). Due to the zonal orientation in summer, any local land-atmosphere feedbacks involving soil moisture or snow heating anomaly can generate a sequence of waves of similar cover. wavelength downstream of the jet, forming a stagnant wave In summer, the reduced pole-to-equator temperature gradient packet that affects weather conditions far away. Recent research (as compared to winter) leads to weaker and more-narrow upper- has shown that when synoptic-scale waves (wavenumbers 6–8) level westerlies, and the stratospheric polar vortex is absent all are trapped in a (near) circumglobal waveguide, wave-resonance 19,64–66 together (Fig. 2). The role of the stratosphere in influencing can greatly increase their amplitude . The wave-resonance boreal summer weather is therefore considered non-existent . mechanism can lead to highly persistent and anomalous weather Moreover, variations in tropical SST have less influence on mid- conditions around the hemisphere and studies have linked it to latitude circulation in summer compared to winter (e.g., several recent high-impact summer extremes, including heat- 38,39 57,66 refs. ). The position of abnormally warm SST in the tropics waves and floods . Though there is a solid theoretical basis determines where the strongest deep convection takes place underlying wave-resonances, their exact significance in the real- associated with shifts in the Walker circulation. The upper-level world in causing extreme weather events is debated . latent heat release during deep convection can trigger long Rossby Finally, the Arctic frontal zone that develops around 70° N in waves that propagate poleward and influence mid-latitude summer is likely to be affected by AA. The warm, snow-free weather . This mechanism is less important in summer than land surface and the cold Arctic Ocean create a strong thermal winter for two reasons: First, the El Niño-Southern Oscillation contrast along the Arctic seaboard around 70° N, generating (ENSO), the dominant mode of variability in tropical SST, tends strong westerlies here. These sub-polar westerlies, together with to peak in boreal winter and is much weaker during boreal the pronounced sub-tropical jet, form the distinct summer feature summer . Second, the prevailing easterly winds in the tropics in of double jets (Fig. 2). A double-jet regime is characterized by a summer limit the ability of Rossby waves to propagate pole- very confined sub-tropical jet with sharp edges wherein wind ward . This is not to say that the tropics cannot influence speeds change rapidly with latitude. Such sharp sub-tropical jets summer mid-latitude weather. Among other things, ENSO gives and thus double jets favor waveguide are effective waveguides 42,43 69 some predictive seasonal forecast skill in summer and the formation and wave-resonance events . Interactions of the two varying location and intensity of monsoon systems, notably over jets can produce high-amplitude atmospheric waves, creating the 66,70 South and East Asia, can affect the mid-latitude summer circu- deepening of troughs and stagnation of ridges . 44–46 lation . Given these specific characteristics of the summer circulations, The state of the cryosphere, in terms of sea-ice and snow cover, several mechanisms have been proposed that link AA with from late winter to early summer can influence the strength and summer mid-latitude weather patterns. These are grouped into latitude of the summer time jet . Boreal snow cover during weakening of the storm tracks, shift in the latitudinal position of spring and summer has shrunk dramatically in recent years, even the mid-latitude jet, and amplification of circumglobal wave 48,49 faster than the decline of Arctic sea-ice extent . Without snow, trains (Fig. 3). the surface albedo is lower and thus the land regions absorb more incoming solar radiation. Furthermore, declining snow cover in spring has a delayed drying effect on the soils by mid-summer, Influence of the Arctic on summer circulation favoring enhanced temperatures due to suppressed evaporative Weakening storm tracks. Theoretical, observational and mod- 50,51 cooling . These thermodynamic processes in conjunction with eling evidence supports the hypothesis that summer storm tracks 26,71,72 reductions in early season snow cover can affect regional to weaken with enhanced Arctic warming . The theoretical 52,53 hemispheric circulation . basis underlying AA and resultant weakening of the mid-latitude The waveguide effect, i.e., the trapping and focusing effects of storm track is straightforward: The thermal-wind balance relates the seasonal jet streams on low-frequency tropospheric waves, vertical shear in the westerly flow to the magnitude of the pole- has an important role in how a changing mid-latitude circu- ward temperature gradient. In the lower troposphere, a reduction 39,54 lation might promote stagnant weather patterns .Wave- in the temperature gradient equates to a similar reduction in the guides produce zonally oriented chains of perturbations that shear, weakening the thermally driven jet and reducing the low- fluctuate at relatively low frequency ranging from weeks to level baroclinicity . A reduced low-level baroclinicity implies less months, creating a teleconnection pattern. Early research found or weaker synoptic-scale cyclogenesis and thus leads to overall that atmospheric disturbances near the jet core are refracted weakening of the storm tracks. Note, that the thermal-wind toward the core, meaning that the jet acts as a waveguide .The balance does not give a direction of causality per se: The causality energy of the trapped disturbances does not disperse strongly could be the other way around, whereby a change in mid-latitude and therefore can propagate much further and possibly circulation alters the poleward heat transport giving rise to more become circumglobal. Such CGWT can generate heatwaves rapid warming in the Arctic . 4 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications P NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE In past climate In warming climate AA Weakening of storm tracks Storm systems Relatively warm Arctic Storm track Relatively cool Arctic Equatorward shift of polar jet AA vs. poleward shift of subtropical jet Amplification of quasi- AA stationary short-wave trains Arctic circle warming over land Fig. 3 Schematic representation of proposed dynamical mechanisms in summer. a Weakening of storm tracks, b latitudinal-shift in jet positions, and c amplification of quasi-stationary waves (Figure created using Apple’s Keynote software) Empirical evidence based on multiple datasets shows that over the satellite-covered period (i.e., since 1979), the mid-latitude summer circulation has indeed weakened in conjunction with a reduction in the poleward temperature gradient in the lower troposphere. This weakening has been detected in the westerly jet (following the thermal-wind balance), the total kinetic energy of –5 synoptic storm systems (by about 15%) and the number of strong 26,71,74 cyclones . Similarly, strong Arctic sea-ice melting years are 75 –10 characterized by a weakened circulation . While, the satellite era is most reliable when analyzing wind field characteristics, its limited timespan compromises long-term trend analyses. Natural –15 variability on multi-decadal time scales, either due to changes in SSTs or from internal atmospheric variability, are thus likely to –20 have a role in the observed trends. There is modeling evidence indicating that these observed trends are at least partly attributable to AA. CMIP5 coupled –25 model simulations of the twentieth century show that the observed changes in the zonal-mean temperature gradient in Slope = 1.4 –30 summer (characterized by AA and enhanced high-latitude land p val = 3.5e–05 warming) are likely attributable to anthropogenic forcing ("likely" according to IPCC lexicon) . Idealized modeling experiments –10 –5 0 5 support storm track weakening when sea-ice is reduced but also Δ U (%) indicate that sea-ice changes by itself can explain only part of the Fig. 4 Observed and projected changes in the mid-latitude Northern observed weakening . Modeling studies indicate that the effects Hemisphere summer storm tracks and westerlies. The percentage of historic sea-ice reductions can explain up to one-third of the change in summer storm tracks (vertical axis) and westerlies (horizontal magnitude of the observed anomalies, with an additional role for 72,78 axis) in future (2081–2100, under scenario RCP8.5) relative to changes in SSTs . Thus, other factors including natural 1981–2000 for individual CMIP5 climate models is shown, and their variability likely had a role in the recently observed summer linear fit (solid black line). Observed changes based on ERA-Interim data circulation changes, but a substantial share of it is likely are given for the 1979–2013 period. Taken from (Coumou et al. .) attributable to AA. NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 5 Δ EKE (%) REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 6 2 –1 10 m s 1.6 1.2 0.8 0.4 –0.4 –0.8 –1.2 –1.6 –2 Fig. 5 Enhanced circumglobal wave train embedded in the summer jet. Linear trends from 1979 to 2010 in the July 250 hPa stream function in the short- wave regime (blue-red shading) computed with the long wavenumbers (1–4) removed. The change in the short waves is embedded in the climatological July-mean 250-hPa wind speed depicting the jet stream (black contour lines). Adapted from (Wang et al. .) 90,91 For future high-emission scenarios, models robustly project observations of clouds and is most significant in winter. storm track weakening, supporting the hypothesis that AA is This is thus largely in agreement with model projections pointing associated with weakened summer storm tracks (see Fig. 4) at the important role of the tropics in shifts in the jet position. 26,71,79,80 . The changes at the end of the century in the high- emission scenario are comparable to the observed changes over Amplification of wave trains. Limited evidence from theory, the past decades . This suggests that either the models under- observations and some model simulations suggests that AA may estimate the future changes (models also underestimate historic amplify synoptic-scale, quasi-stationary waves embedded in the changes in the Arctic itself) or that a substantial part of the summer jet (Fig. 5). Theory of the dynamics of a dry atmosphere observed trend is associated with multi-decadal natural varia- suggests that a lower troposphere diabatic heating source in the bility . For the North American sector, the amount of AA in mid-latitudes will have a larger stationary wave response (in models by the end of century is negatively correlated with the terms of a meridional stream function displacement) when the changes in jet speed and wave phase speed in summer . Models background baroclinicity and zonal winds are reduced, as a direct robustly project a weakening of summer storm tracks by the end response of AA . In a more complex atmosphere of an aqua- of the century (Fig. 4), but this is not the case for the upper-level planet model (i.e., an Earth covered by water only), quasi- jet: There is large inter-model spread with some models stationary synoptic-scale wave trains are enhanced when the projecting a strengthening and some a weakening of the upper- meridional temperature gradient is reduced and the westerly 26,28 level jet (Fig. 4) . winds weaken . Thus, as the background flow becomes weaker, the same heating source in the lower troposphere can trigger a Shift in jet-position. Understanding jet shifts in the Northern stronger stationary wave response especially for synoptic-scale Hemisphere is a challenging task due to several competing pro- waves. The increased moisture content in a warmer atmosphere, cesses. Theoretically, the change in jet stream position can be and the tendency for increased latent heat release in the tropics divided into a relatively small direct radiative-induced shift and a and over warm ocean currents in higher latitudes, can provide larger indirect SST-mediated effect. The indirect effect comes further heating to perturb more or stronger CGWTs. from mid-latitude dynamical feedbacks, involving enhanced Some observational evidence suggests that the quasi-stationary irreversible mixing due to wave breaking of high-frequency component of mid-latitude summer circulation has become transient eddies , and can explain most of the expected jet wavier since 1979, in particular over the North American 83 60,62,93,94 shifts . When considered in isolation, AA should theoretically sector . Figure 5 plots linear trends over 1979–2010 in cause a southward shift in the mid-latitude jet stream. the short-wave regime showing enhanced CGWTs over both the Idealized dry atmospheric model simulations indeed indicate a North American and Eurasian sectors. For the American sector, southward shifted jet coming from AA by itself . This is also this is further supported by detected increases in waviness 60,62 confirmed by more complex models with reductions in sea-ice metrics (see Box 3). Using three different climate models, the 31,72,78,85 imposed . Despite the process linking AA to a more summer time amplification of quasi-stationary short waves over equatorward jet, the zonal-mean jet streams are projected to the American sector appears to be attributable to greenhouse gas migrate poleward by about one degree by the end of the twenty- forcing . As the CGWT is linked to the summer North Atlantic 27,86 39 first century under a high-emission scenario . Thus in the long Oscillation (SNAO, Box 2) , an AA-like circulation anomaly run, i.e., at the end-of-century, the tropics likely dominate the (i.e., negative phase of the SNAO) could modulate the CGWT tug-of-war, at least in models. While in winter both the Atlantic and vice versa. However, the summer CGWT can also be and the Pacific jets are projected to migrate poleward, in summer triggered by heating sources associated with the Indian 28 44 this shift is seen only for the North Atlantic jet . AA may monsoon . therefore exert a stronger opposing influence to the expected The few modeling studies with historically observed low sea-ice poleward shift of the Pacific jet in summer. Different state-of-the- concentration show a stationary wave-train response in summer 78,95 art climate models employ different simplifying parameteriza- emerging from AA . Still, future model projections of mid- tions (e.g., for clouds) and those can, in a complex, non-linear latitude quasi-stationary short-wave patterns are generally system, lead to very different outcomes . The inter-model spread inconsistent with recent observations. Future CMIP5 projections of the poleward shift therefore tends to be larger than the signal under high-emission scenarios show an overall decrease in 27,46,86 96 itself . blocking both for winter and summer . The reasons behind these Observed jet shifts in the Northern Hemisphere are generally divergent findings are not well understood and may result from small compared to those in the Southern Hemisphere, but still competing effects from the tropical monsoons , changes in land- 88 46 generally indicate a poleward migration . For the Northern sea thermal contrast , model biases in representing summer time Hemisphere, some evidence of a poleward shift of jet streams Rossby waves and the use of different diagnostics to quantify 88,89 has been identified in reanalysis products and satellite waviness (see Box 3). 6 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE BOX 2 | Atmospheric teleconnections in summer Remote climate effects, known as atmospheric teleconnections, consist of three main types: (i) regional long-wave pattern confined in a longitudinal sector of the globe, like the Pacific-North America pattern and the North Atlantic Oscillation ; (ii) a hemisphere-wide pattern with a prominent zonal- 70,136,137 mean component like the Arctic Oscillation or the Annular Modes ; and (iii) the trapping and focusing effects of the seasonal jet streams on 39 54 low-frequency tropospheric waves , known as the waveguide effect . The figure below shows schematic representations of these three types of teleconnection patterns. The summer North Atlantic Oscillation (SNAO) can be regarded as the counterpart of the more robust winter NAO. The centers of action of the SNAO exhibit a more northerly location, have a smaller geographical extent and a weaker dipole pattern compared to the winter NAO. Panel a schematically shows the surface pressure patterns of the positive SNAO phase, i.e., when pressure differences are strong, together with the position of the warmer/ drier and colder/wetter regions. Generally, the negative phase shows a reverse pattern. Like the winter NAO, the impact of the SNAO on climate extremes such as heavy rainfall and flooding is profound, especially for Europe. Post-2007 summers have seen increasingly robust negative SNAO associated with a persistent anticyclonic anomaly over Greenland and a cyclonic anomaly over Northwest Europe. This pattern caused rapid melting of the Greenland ice sheet and brought unusually wet summers to Northwest Europe, including the massive flooding of U.K. in summer 2012. Future projections of climate models suggest an increasingly positive SNAO in warmer climates . The Northern Hemisphere Annular Mode (NAM) is an internally driven atmospheric mode maintained by both stationary and transient waves. The NAM is defined as the first EOF of monthly 500 hPa geopotential height fields and coincides with the definition of the Arctic Oscillation during the winter months . Winter and summer NAM patterns present both a diverse geopotential height field, mean meridional circulation and eddy structure. The summer NAM has a smaller latitudinal extent and has a stronger link to surface air temperature over Eurasia. The positive phase of the summer NAM is associated with negative geopotential height anomalies over Greenland and the Arctic Ocean and an annual band of positive anomalies comprised between 40° and 60° N and particularly extended over Eurasia, as schematically shown in panel b. When the summer NAM is strongly positive, the storm tracks follow the Arctic front . Anomalously positive summer NAM phases are associated with double jets favoring blocking between the polar and the subtropical jets. During a positive summer NAM phase, surface temperatures over Eurasia show a dipole pattern with warmer conditions over Europe and colder conditions over East Asia . The energy of waves trapped in a waveguide is not dispersed as broadly as in teleconnections of type (i) and therefore it can propagate farther before being dissipated. In summer, when an efficiently trapping waveguide becomes (almost) circumglobal, then resonant interactions between free and forced waves (typically of synoptic scale, wavenumbers 6–8) might lead to wave-amplification and persistent, high-amplitude waves. Waveguide formation is favored during double-jet regimes and thus linked to strongly positive NAM phases. Panel c shows a schematic representation of a wave- resonance event with an amplified wave 7. The orange and green areas represent regions of positive and negative upper-level meridional winds. Such events are associated with alternating hot-dry and cold-wet conditions, following the ridges and troughs. Such a situation is prone to blocking weather systems and deepened troughs, and their relative longevity is key to making severe weather extremes . Wave-resonance periods have therefore been 57,64 linked to both persistent heatwaves and severe flooding events . Finally, blocking itself is also distinctly different in summer compared to winter. Upstream latent heat release has been identified as an important contributor to persistent blocking and this mechanism is especially important in summer . a c SNAO well above zero Summer NAM positive Wave-resonance Colder/wetter Storm tracks Warmer/drier Figure Box 2: Panel a shows a schematic of the positive SNAO indicating the anomalously low (blue) and high (red) sea level pressure regions together with cold/wet and dry/warm regions. Panel b shows a schematic of the 500 hPa geopotential height configuration during the positive phase of the NAM. Rainy clouds mark the position of the storm tracks during a strongly positive NAM phase. Panel c shows a schematic of an amplified wave 7. Green and orange regions show the position of pronounced northward and southward wind anomalies. (Figure created using Python, GIMP, Powerpoint and Inkscape software) Double jets reduction of snow cover. Thus, while the overall equator-to-pole Basic theory suggests that double-jet flow regimes are to become temperature gradient reduces, the thermal gradient actually slightly more common with pronounced AA . Such flow regimes increases at the land-ocean boundary around the Arctic are characterized by sharper sub-tropical jets (i.e., a strong mer- circle. This situation favors the formation of the Arctic front jet at 68,69 idional wind shear) which can act as waveguides . Moreover, ~70° N in addition to the sub-tropical jet which is normally in summer, recent AA is characterized by enhanced land present. Such double-jet regimes have become more frequent in warming over the high latitudes (~70° N) and much less warming recent years due to high-latitude land warming, something which over the nearby Arctic Ocean, in stark contrast to winter warming is partly attributable to anthropogenic greenhouse gas forcing . patterns. This enhanced high-latitude land warming is likely Double jets favor waveguide formation and wave-resonance but related to a combination of the smaller heat capacity of land the evidence for an increase in frequency or persistence of 66,69 compared to ocean, as well as late-spring to early-summer waveguides is limited . NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 7 REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 BOX 3: | Measuring mid-latitude waviness Although the concept of wavy versus zonal circulation patterns is straightforward, quantifying these opposite states has proven challenging. Approaches can be roughly separated into geometric and dynamic methods. The former focuses on the geometry of the circulation to characterize the 61,62,140,141 departure of the flow from zonality in terms of wave amplitude, sinuosity, or circularity . These metrics have the advantage of being intuitive and readily visualized from geopotential height contours, but they have been criticized for lacking a firm physical basis. The existence of wave trains usually lasts 2–4 weeks and they do not necessarily have a preferred phase position . Averaging over time may thus cancel out any signal. Further, 58,64,65 often the diagnostic approach to detect extratropical jet’s waviness involves Fourier decomposition of zonal wavenumbers which could lead to spurious wave signals. In the extreme case: Fourier decomposition of a delta function will create amplitudes in a range of zonal wavenumbers even if the circulation feature is extremely local, i.e., it does not involve a wave “train” . Likewise, a zonal wavenumber 5 along 40° N (subtropical jet) or 65° N (polar jet) depicts two very different wavelengths. Dynamically based waviness metrics such as effective diffusivity of potential vorticity and finite- 143,144 amplitude wave activity are derived from first-order mass and circulation conservation principles and satisfy exact budget closure equations. These measures provide a theoretical basis for quantitatively relating changes in zonal wind speed to accompanying changes in wave amplitude and eddy fluxes, which is straightforwardly verifiable under idealized conditions. Such approaches are being applied in climatological studies of circulation trends and extreme weather events related to amplified flow patterns but their derivation is more technical and their application is more involved than geometric methods. Combined effects The Northern Hemisphere summer circulation is connected to The described dynamical mechanisms do not operate in isolation the tropics primarily via two-way interactions with monsoon but instead interact with each other, with other teleconnections, systems. Anomalously low snow cover over Eurasia during late and regionally with land-atmosphere feedbacks. The exact nature winter to early-spring can increases the land-ocean temperature of these interactions is still speculative at this stage but positive gradient due to snow-albedo and soil moisture effects which feedbacks are certainly possible which would lead to tail risks. strengthens the Indian summer monsoon rainfall . On sub- Weakened storm tracks favor the buildup of hot and dry seasonal time scales, extreme monsoonal rainfall over eastern or conditions over the continents. This can strengthen the land- western Himalayan foothills is often linked to southward ocean thermal contrast which, combined with the enhanced wave intruding mid-latitude troughs, possibly part of the CGWT . response to thermal forcing when the background flow is weak, The monsoon is also a source of diabatic heating that can gen- could lead to amplified quasi-stationary waves. The projected erate the CGWT . Thus, wave trains originating from higher weakening of storm tracks and westerlies can interact with soil latitudes can modulate the intensity of the monsoon, and the moisture and snow cover changes, potentially regionally exacer- monsoon’s strength in turn reinforces the downstream propaga- bating zonal-mean circulation changes. Enhanced terrestrial tion of the wave train . The projected increase in Indian summer heating in mid- to high-latitudes associated with reductions in monsoon rainfall is thus expected to strengthen the CGWT. snow cover can promote anomalous ridging in the upper tropo- sphere, as suggested by both observational and modeling Robust evidence and knowledge gaps 53,97,98 studies . The emergence of such positive geopotential height Several arguments support the AA influence on summer circu- anomalies in mid-to-high latitudes induces hot and dry air lation as compared to winter. First, in the mid-latitudes, the through subsidence and easterly wind anomalies of continental equator-to-pole near-surface temperature gradient has seen a origin . Likewise, drying soils in subtropical regions like the pronounced reduction in summer, but not in winter . Second, Mediterranean (a robust projection in future climate) can favor late-spring to early-summer snow cover extent, which influences 48,109 the formation of a heat low, i.e., low pressures at the surface due summer flow regimes, has dramatically declined ; and finally, to upflow associated with intense surface heating . A Medi- summer circulation is less affected by tropical ENSO forcing, and 17,38,44 terranean near-surface heat low brings easterly winds to central thus potentially more sensitive to Arctic warming . and western Europe, obstructing the normal westerlies . Wea- Reviewing the literature, we conclude that there is robust evi- kened or diverted westerlies can reduce rainfall leading to further dence that AA causes a weakening of storm tracks in summer. soil drying. Due to such feedback mechanisms between soil The mechanism is straightforward: Cyclone genesis is directly moisture, snow cover changes, and continental-scale circulation, related to the lower troposphere temperature gradient which summer weather in western Europe and interior North America weakens with AA. Further support comes from observed trends 60,100,101 is likely to become more continental as seen in models and historic and future climate modeling experiments. Still, 24,25,102 and possibly also in observations . The effectiveness of this multi-model attribution analyses are needed to quantify the exact mechanism depends on the delicate balance between land- role of past and future AA on the weakening of storm tracks. atmosphere feedbacks and the strength (and thus its long-term There is still substantial uncertainty in what implications this weakening) of the storm tracks. weakening will have for summer weather conditions. Does it Recent studies indicate that the Atlantic Meridional Over- become more persistent and therefore more extreme? Theoreti- turning Circulation (AMOC, i.e., the large-scale north-south cally, a weakened westerly flow leads to a small increase in the transport in the Atlantic ocean) has seen an unprecedented stationary wavenumber (i.e., a 5% decrease in the flow would lead 103,104 slowdown in recent decades , something which is projected to a 2.5% increase in stationary wavenumber ) but not neces- for future warmer climates as well. This slowdown results in sarily to a more-stationary flow. Both observations and climate anomalously cold SSTs over the northern Atlantic which can models suggest that a weaker westerly flow is associated with a trigger a quasi-stationary Rossby wave response favoring blocking more wavy flow pattern, especially in summer, but this does not 60,61 high-pressure systems over western Europe . So, just like AA, a necessarily imply a cause-effect relationship (see Box 3). A slowdown of the AMOC leads to weakening westerlies in summer reduction in synoptic activity favors the buildup of hot-dry over the Atlantic sector, favoring persistent hot-dry extremes over conditions and this can interact with regional land-atmosphere Europe . Recent observational studies indeed indicate that processes that can feedback on the continental-scale circulation weather persistence in Europe and some other mid-latitude (as outlined above). However, the relative role of these processes, 23–25,102 regions has increased in boreal summer . and the strength of their interactions is largely unquantified. To 8 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE assess future high-end risks, these non-linear interactions need to the atmosphere. This can give direct insight into the underlying be disentangled and quantified. physical reasons behind model bias and thus provides concrete There is robust evidence that summer time quasi-stationary targets for model improvements. waves lead to persistent and therefore extreme weather condi- Finally, high-resolution paleo-climate records over the Holo- tions. However, substantial uncertainty remains in how such cene period can provide further insights into the circulation waves will change under global warming including the role of AA response to temperature gradient changes and put recent trends therein. While physical mechanisms exist that could amplify into a long-term perspective. Paleo proxies typically measure quasi-stationary short waves as an indirect response to AA, their some form of biological activity and are therefore best suited to representation in climate models is biased and their relative analyze the summer/growing season. The mid-Holocene provides importance compared to other drivers is poorly understood . a possible paleo analog with enhanced high-latitude warming Some upward trends in quasi-stationary wave activity have been and, interestingly, this period was also characterized by enhanced reported but confidence in these trends is generally low due to the drought conditions in the mid-latitudes . Likewise, tree-ring use of different diagnostics leading to conflicting results (Box 3). analyses suggest that enhanced jet stream waviness during sum- Moreover, the relative importance of enhanced high-latitude mers in recent decades has been unprecedented over the post- warming, strengthened monsoons, and a weakening AMOC with 1725 period . associated mid-latitude SST anomalies, in shaping the mid- latitude summer circulation needs to be understood. All these Summary drivers have a tendency to strengthen CGWTs but their inter- Future impacts from extreme weather are likely to be most pro- actions and regional effects are poorly known. nounced in summer, as most ecological activity and agricultural production takes place in this season . Though the uncertainties are large, changes in atmosphere dynamics have the potential to Ways forward cause rapid transitions at a regional scale leading to surprises for Given the societal risks and large uncertainties, we argue for society. In summer synergistic effects between thermodynamic coordinated research efforts to address the knowledge gaps and dynamic drivers of extreme weather could act in the same described above. To efficiently address this overarching theme, direction to cause very-extreme extremes . Recent summers have tighter collaboration between sub-disciplines within climate sci- seen such anomalous weather (Box 1) and these events are not ences is desirable, including scientists studying Arctic processes, well understood. This presents risks for society and in particular monsoons, storm track dynamics, (sub) seasonal forecasts, and for global food production, given that the major breadbasket extreme weather. regions are located in the mid-latitudes with many crop types Idealized atmosphere models are useful to study individual tel- vulnerable to heat extremes . econnections and their drivers. However, a central challenge is to The current literature provides robust evidence that AA quantify the interactions between Arctic teleconnections and other influences mid-latitude summer circulation substantially by teleconnections and regional processes, requiring state-of-the-art weakening the storm tracks. The uncertainties to do with other weather or climate models. Teleconnections are generally state- dynamical aspects and with how dynamical changes ultimately dependent (e.g., the Arctic’sinfluence might be pronounced only if affect regional weather conditions are admittedly large. thetropics areinaspecific state) and non-stationary. While often Nevertheless, we identified several possible feedback mechan- the mean circulation response to a certain perturbation (e.g., isms for how storm track weakening can lead to persistent and removing sea-ice) is analyzed, it will be important to quantify the therefore extreme weather in the mid-latitudes. Several studies response in probabilistic terms: The change in frequency of certain suggest that Northern Hemisphere summer weather is indeed 23,24,66,119 high-impact circulation regimes such as amplified and persistent already becoming more persistent . quasi-stationary waves. An increased frequency in such rare regimes In summary, this review shows that AA is likely to have sub- will have little influence on the mean circulation but can have large stantial impacts on mid-latitude summer circulation. The societal societal impacts. To do so will require large ensembles and well- impacts can be severe due to tail risks arising from radiatively coordinated experiments with a range of different models. To dis- forced mean summer warming combined with local and remote entangle the influence of different teleconnections, a so-called processes that favor more persistent summer weather. A coor- storyline approach can be insightful as well. This avoids quantifying dinated research agenda focusing on summer circulation, its probabilities associated with dynamical changes altogether and, drivers and extremes is needed to resolve the key knowledge gaps. instead, creates a set of physically plausible scenarios (i.e., storylines) 110,111 of future changes . For example, the change in European Received: 26 July 2017 Accepted: 20 June 2018 summer circulation could be described by combining a set of storylines that are based on the response to remote drivers from the Arctic, the tropics, AMOC, and regional soil-moisture changes. This way, the high-end risk (e.g., when Arctic and tropical teleconnec- tions combine with soil-moisture feedbacks to push European summer weather towards much more persistent hot-dry condi- References 1. Lehmann, J., Coumou, D. & Frieler, K. Increased record-breaking tions) can be identified and studied without assigning a specific precipitation events under global warming. Clim. Chang. 132, 501–515 (2015). probability to it . 2. Westra, S., Alexander, L. V. & Zwiers, F. W. Global increasing trends in To understand and overcome model biases, e.g., in the repre- annual maximum daily precipitation. J. Clim. 26, 3904–3918 (2013). sentation of summer Rossby waves and ocean-atmosphere 3. Fischer, E. M. & Knutti, R. Anthropogenic contribution to global occurrence feedbacks in the presence of sea-ice , novel machine learning of heavy-precipitation and high-temperature extremes. Nat. Clim. Chang. 5, approaches should be used to better integrate information from 560–564 (2015). 4. Coumou, D. & Rahmstorf, S. A decade of weather extremes. Nat. Clim. Chang. observations in climate models. In particular, causal discovery 2, 491–496 (2012). algorithms can identify causal pathways in the atmosphere and 5. Shepherd, T. G. Atmospheric circulation as a source of uncertainty in climate quantify their relative importance from observations alone . change projections. Nat. Geosci. 7, 703–708 (2014). They can thus be used to do process-based model validations and 6. Cohen, J. et al. Recent Arctic amplification and extreme mid-latitude weather. quantify how well models represent certain causal pathways in Nat. Geosci. 7, 627–637 (2014). NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 9 REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 7. Screen, J. A. Climate science: far-flung effects of Arctic warming. Nat. Geosci. 38. Schubert, S., Wang, H. & Suarez, M. Warm season subseasonal variability and 10, 253–254 (2017). climate extremes in the Northern Hemisphere: The role of stationary Rossby 8. Vihma, T. Weather extremes linked to interaction of the Arctic and waves. J. Clim. 24, 4773–4792 (2011). Midlatitudes. Clim. Extremes 226,39–50 (2017). 39. Branstator, G. Circumglobal teleconnections, the jet stream waveguide, and 9. Kretschmer, M., Coumou, D., Tziperman, E. & Cohen, J. More-persistent the North Atlantic oscillation. J. Clim. 15, 1893–1910 (2002). weak stratospheric polar vortex states linked to cold extremes. Bull. Am. 40. Trenberth, K. E. et al. Progress during TOGA in understanding and modeling Meteor. Soc. 99,49–60 (2018). global teleconnections associated with tropical sea surface temperatures. 10. Budikova, D. Role of Arctic sea ice in global atmospheric circulation: a review. J. Geophys. Res. 103, 14291–14324 (1998). Glob. Planet. Chang. 68, 149–163 (2009). 41. Yu, J.-Y., Wang, X., Yang, S., Paek, H. & Chen, M. The changing El Nino- 11. Overland, J. E. et al. Nonlinear response of mid-latitude weather to the Southern oscillation and associated climate extremes. Clim. Extremes 226, changing Arctic. Nat. Clim. Chang. 6, 992–999 (2016). 3–38 (2017). 12. Walsh, J. E. Intensified warming of the Arctic: causes and impacts on middle 42. Eden, J. M., Oldenborgh, G. J., Van, Hawkins, E. & Suckling, E. B. A global latitudes. Glob. Planet. Chang. 117,52–63 (2014). empirical system for probabilistic seasonal climate. Geosci. Model Dev. 8, 13. Francis, J. A., Vavrus, S. J. & Cohen, J. Amplified Arctic warming and mid- 3947–3973 (2015). latitude weather: new perspectives on emerging connections. WIREs Clim. 43. Weisheimer, A. & Palmer, T. N. On the reliability of seasonal climate Chang. 8, e474 (2017). forecasts. J. R. Soc. Interface 11, 20131162 (2014). 14. Vihma, T. Effects of Arctic sea ice decline on weather and climate: a review. 44. Ding, Q. & Wang, B. Circumglobal teleconnection in the Northern Surv. Geophys. 35, 1175–1214 (2014). Hemisphere Summer*. J. Clim. 18, 3483–3505 (2005). 15. Shepherd, T. G. Effects of a warming Arctic. Science 353, 989–990 (2016). 45. Ding, Q. & Wang, B. Intraseasonal teleconnection between the summer 16. Cohen, J. et al. Arctic change and possible influence on mid-latitude climate Eurasian wave train and the Indian Monsoon. J. Clim. 20, 3751–3767 and weather. US CLIVAR White Paper. https://doi.org/10.5065/ (2007). D6TH8KGW (2018). 46. Shaw, T. A. & Voigt, A. Tug of war on summertime circulation between 17. Horton, R. M., Mankin, J. S., Lesk, C., Coffel, E. & Raymond, C. A review of radiative forcing and sea surface warming. Nat. Geosci. 8, 560–566 (2015). recent advances in research on extreme heat events. Curr. Clim. Chang. Rep. 2, 47. Hall, R. J., Jones, J. M., Hanna, E. & Scaife, A. A. Drivers and potential 242–259 (2016). predictability of summer time North Atlantic polar front jet variability. Clim. 18. Schär, C. et al. The role of increasing temperature variability in European Dyn. 48, 3869–3887 (2017). summer heatwaves. Nature 427, 332–336 (2004). 48. Derksen, C. & Brown, R. Spring snow cover extent reductions in the 2008- 19. Petoukhov, V., Rahmstorf, S., Petri, S. & Schellnhuber, H. J. Quasiresonant 2012 period exceeding climate model projections. Geophys. Res. Lett. 39, amplification of planetary waves and recent Northern Hemisphere weather L19504 (2012). extremes. Proc. Natl Acad. Sci. USA 110, 5336–5341 (2013). 49. Kunkel, K. E. et al. Trends and extremes in Northern Hemisphere snow 20. Russo, S. et al. Magnitude of extreme heat waves in present climate and their characteristics. Curr. Clim. Chang. Rep. 2,65–73 (2016). projection in a warming world. J. Geophys. Res. Atmos. 119,500–12,512 (2014). 50. Wu, R., Zhao, P. & Liu, G. Change in the contribution of spring snow cover 21. Christidis, N., Jones, G. S. & Stott, P. A. Dramatically increasing chance of and remote oceans to summer air temperature anomaly over Northeast China extremely hot summers since the 2003 European heatwave. Nat. Clim. Chang. around 1990. J. Geophys. Res. Atmos. 119, 663–676 (2014). 5,46–50 (2014). 51. Halder, S. & Dirmeyer, P. A. Relation of Eurasian snow cover and Indian 22. Weaver, S. J., Kumar, A. & Chen, M. in Climate Extremes: Patterns and summer monsoon rainfall: Importance of the delayed hydrological effect. Mechanisms, Geophysical Monograph 226 1st edn (eds Simon Wang, S.-Y. J. Clim. 30, 1273–1289 (2017). et al.) 105–114 (Wiley, New York, 2017). 52. Alexander, M., Tomas, R., Deser, C. & Lawrence, D. The atmospheric 23. Pfleiderer, P. & Coumou, D. Quantification of temperature persistence over response to projected terrestrial snow changes in the late twenty-first century. the Northern Hemisphere land-area. Clim. Dyn. 51, 627–637 (2018). J. Clim. 23, 6430–6437 (2010). 24. Hoffmann, P. Enhanced seasonal predictability of the summer mean 53. Matsumura, S. & Yamazaki, K. Eurasian Subarctic summer climate in temperature in Central Europe caused by new dominant weather patterns. response to anomalous snow cover. J. Clim. 25, 1305–1317 (2012). Clim. Dyn. 50, 2799–2812 (2018). 54. Hoskins, B. J. & Karoly, D. J. The steady linear response of a spherical 25. Horton, D. E. et al. Contribution of changes in atmospheric circulation atmosphere to thermal and orographic forcing. J. Atmos. Sci. 38, 1179–1196 patterns to extreme temperature trends. Nature 522, 465–469 (2015). (1981). 26. Coumou, D., Lehmann, J. & Beckmann, J. The weakening summer circulation 55. Hoskins, B. J. & Ambrizzi, T. Rossby wave propagation on a realistic in the Northern Hemisphere mid-latitudes. Science 348, 324–327 (2015). longitudinally varying flow. J. Atmos. Sci. 50, 1661–1671 (1983). 27. Barnes, E. A. & Polvani, L. Response of the midlatitude jets, and of their 56. Teng, H., Branstator, G., Wang, H., Meehl, G. A. & Washington, W. M. variability, to increased greenhouse gases in the CMIP5 models. J. Clim. 26, Probability of US heat waves affected by a subseasonal planetary wave pattern. 7117–7135 (2013). Nat. Geosci. 6, 1056–1061 (2013). 28. Barnes, E. A. & Polvani, L. M. CMIP5 projections of Arctic amplification, of 57. Stadtherr, L., Coumou, D., Petoukhov, V., Petri, S. & Rahmstorf, S. Record the North American/North Atlantic circulation, and of their relationship. Balkan floods of 2014 linked to planetary wave resonance. Sci. Adv. 2, J. Clim. 28, 5254–5271 (2015). e1501428 (2016). 29. Harvey, B. J., Shaffrey, L. C. & Woollings, T. J. Equator-to-pole temperature 58. Wang, S. Y., Hipps, L., Gillies, R. R. & Yoon, J. H. Probable causes of the differences and the extra-tropical storm track responses of the CMIP5 climate abnormal ridge accompanying the 2013–2014 California drought: ENSO models. Clim. Dyn. 43, 1171–1182 (2014). precursor and anthropogenic warming footprint. Geophys. Res. Lett. 41, 30. Harvey, B. J., Shaffrey, L. C. & Woollings, T. J. Deconstructing the climate 3220–3226 (2014). change response of the Northern Hemisphere wintertime storm tracks. Clim. 59. McKinnon, K. A., Rhines, A., Tingley, M. P. & Huybers, P. Long-lead Dyn. 45, 2847–2860 (2015). prediction of eastern United States hot days from Pacific sea surface 31. Zappa, G., Pithan, F. & Shepherd, T. G. Multi-model evidence for an temperatures. Nat. Geosci. 9, 389–394 (2016). atmospheric circulation response to Arctic sea ice loss in the CMIP5 future 60. Vavrus, S. J. et al. Changes in North American atmospheric circulation and projections. Geophys. Res. Lett. 45, 1011–1019 (2018). extreme weather: influence of Arctic amplification and Northern Hemisphere 32. Perlwitz, J., Hoerling, M. & Dole, R. Arctic tropospheric warming: causes and snow cover. J. Clim. 30, 4317–4333 (2017). linkages to lower latitudes. J. Clim. 28, 2154–2167 (2015). 61. Cattiaux, J., Peings, Y., Saint-Martin, D., Trou-Kechout, N. & Vavrus, S. J. 33. Petoukhov, V. & Semenov, V. A. A link between reduced Barents-Kara sea ice Sinuosity of midlatitude atmospheric flow in a warming world. Geophys. Res. and cold winter extremes over northern continents. J. Geophys. Res. 115, Lett. 43, 8259–8268 (2016). D21111 (2010). 62. Di Capua, G. & Coumou, D. Changes in meandering of the Northern 34. Kim, B.-M. et al. Weakening of the stratospheric polar vortex by Arctic sea-ice Hemisphere circulation. Environ. Res. Lett. 11,1–9 (2016). loss. Nat. Commun. 5, 4646 (2014). 63. Hoskins, B. & Woollings, T. Persistent extratropical regimes and climate 35. Jaiser, R., Dethloff, K., Handorf, D., Rinke, A. & Cohen, J. Impact of sea ice extremes. Curr. Clim. Chang. Rep. 1, 115–124 (2015). cover changes on the Northern Hemisphere atmospheric winter circulation. 64. Petoukhov, V. et al. Role of quasiresonant planetary wave dynamics in recent Tellus A 64,1–11 (2012). boreal spring-to-autumn extreme events. Proc. Natl Acad. Sci. USA 113, 36. Kretschmer, M., Coumou, D., Runge, J. & Donges, J. Using causal effect 6862–6867 (2016). networks to analyze different Arctic drivers of mid-latitude winter circulation. 65. Coumou, D., Petoukhov, V., Rahmstorf, S., Petri, S. & Schellnhuber, H. J. J. Clim. 29, 4069–4081 (2016). Quasi-resonant circulation regimes and hemispheric synchronization of 37. Kidston, J. et al. Stratospheric influence on tropospheric jet streams, storm extreme weather in boreal summer. Proc. Natl Acad. Sci. USA 111, tracks and surface weather. Nat. Geosci. 8, 433–440 (2015). 12331–12336 (2014). 10 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE 66. Kornhuber, K., Petoukhov, V., Petri, S., Rahmstorf, S. & Coumou, D. Evidence 97. Vavrus, S. The role of terrestrial snow cover in the climate system. Clim. Dyn. for wave resonance as a key mechanism for generating high-amplitude quasi- 29,73–88 (2007). stationary waves in boreal summer. Clim. Dyn. 49, 1961–1979 (2017). 98. Matsumura, S., Zhang, X. & Yamazaki, K. Summer Arctic atmospheric 67. Reed, R. J. & Kunkel, B. The Arctic circulation in summer. J. Meteorol. 17, circulation response to spring Eurasian snow cover and its possible linkage to 489–506 (1960). accelerated sea ice decrease. J. Clim. 25, 1305–1317 (2014). 68. Manola, I., Selten, F., De Vries, H. & Hazeleger, W. ‘Waveguidability’ of 99. Fischer, E. M., Seneviratne, S. I., Vidale, P. L., Lüthi, D. & Schär, C. Soil idealized jets. J. Geophys. Res. Atmos. 118, 10432–10440 (2013). moisture-atmosphere interactions during the 2003 European summer heat 69. Kornhuber, K. et al. Summertime planetary wave-resonance in the Northern wave. J. Clim. 20, 5081–5099 (2007). and Southern Hemisphere. J. Clim. 30, 6133–6150 (2017). 100. Haarsma, R. J. et al. Drier Mediterranean soils due to greenhouse warming 70. Tachibana, Y., Nakamura, N., Komiya, H. & Takahashi, M. Abrupt evolution bring easterly winds over summertime central Europe. Geophys. Res. Lett. 36, of the summer Northern Hemisphere annular mode and its association with 1–7 (2009). blocking. J. Geophys. Res. 115, D12125 (2010). 101. Bladé, I., Liebmann, B., Fortuny, D. & van Oldenborgh, G. J. Observed and 71. Chang, E. K. M., Ma, C., Zheng, C. & Yau, A. M. W. Observed and projected simulated impacts of the summer NAO in Europe: implications for projected decrease in Northern Hemisphere extratropical cyclone activity in summer and drying in the Mediterranean region. Clim. Dyn. 39,709–727 (2012). its impacts on maximum temperature. Geophys. Res. Lett. 43,2200–2208 (2016). 102. Alvarez-Castro, M. C., Faranda, D. & Yiou, P. Atmospheric dynamics leading 72. Petrie, R. E., Shaffrey, L. C. & Sutton, R. T. Atmospheric response in summer to West European summer hot temperatures since 1851. linked to recent Arctic sea ice loss. Q. J. R. Meteorol. Soc. 141, 2070–2076 (2015). Complexity 2018, 2494509 (2018). 73. Lee, S. A theory of polar amplification from a general circulation perspective. 103. Rahmstorf, S. et al. Exceptional twentieth-century slowdown in Atlantic J. Atmos. Sci., 50,31–43 (2014). Ocean overturning circulation. Nat. Clim. Chang. 5, 475–480 (2015). 74. Coumou, D., Kornhuber, K., Lehmann, J. & Petoukhov, V. Weakened flow 104. Caesar, L., Rahmstorf, S., Robinson, A. G. F. & Saba, V. Observed fingerprint of persistent circulation, and prolonged weather extremes in boreal summer. a weakening Atlantic Ocean overturning circulation. Nature 556, 191–196 Clim. Extremes 226,61–73 (2017). (2018). 75. Knudsen, E., Orsolini, Y., Furevik, T. & Hodges, K. Observed anomalous 105. Duchez, A., Frajka-Williams, E., Josey, S. A., Evans, D. G. & Grist, J. P. Drivers atmospheric circulation in summers of unusual Arctic Sea ice reduction. of exceptionally cold North Atlantic Ocean temperatures and their link to the J. Geophys. Res. 16, 2014 (2014). 2015 European heat wave. Environ. Res. Lett. 11, 074004 (2016). 76. Mann, M. E. et al. Influence of anthropogenic climate change on planetary 106. Haarsma, R. J., Selten, F. M. & Drijfhout, S. S. Decelerating Atlantic wave resonance and extreme weather events. Sci. Rep. 7, 45242 (2017). meridional overturning circulation main cause of future west European 77. Blackport, R. & Kushner, P. J. The transient and equilibrium climate response summer atmospheric circulation changes. Environ. Res. Lett. 10, 94007 (2015). to rapid summertime sea ice loss in CCSM4. J. Clim. 29, 401–417 (2016). 107. Kripalani, R. H., Kulkarni, A. & Sabade, S. S. El Nino southern oscillation, 78. Screen, J. A. Influence of Arctic sea ice on European summer precipitation. Eurasian snow cover and the indian monsoon rainfall. Indian Natl Sci. Acad. Environ. Res. Lett. 8, 044015 (2013). 67, 361–368 (2001). 79. Lehmann, J., Coumou, D., Frieler, K., Eliseev, A. V. & Levermann, A. Future 108. Vellore, R. K. et al. Monsoon—extratropical circulation interactions in changes in extratropical storm tracks and baroclinicity under climate change. Himalayan extreme rainfall. Clim. Dyn. 46, 3517–3546 (2016). Environ. Res. Lett. 9, 084002 (2014). 109. Tang, Q., Zhang, X. & Francis, J. A. Extreme summer weather in northern 80. Brewer, M. C. & Mass, C. F. Projected changes in Western U. S. large-scale mid-latitudes linked to a vanishing cryosphere. Nat. Clim. Chang. 4,45–50 summer synoptic circulations and variability in CMIP5 models. J. Clim. 29, (2013). 5965–5978 (2016). 110. Zappa, G. & Shepherd, T. G. Storylines of atmospheric circulation change for 81. Deser, C., Knutti, R., Solomon, S. & Phillips, A. S. Communication of the role European regional climate impact assessment. J. Clim. 30, 6561–6577 (2017). of natural variability in future North American climate. Nat. Clim. Chang. 2, 111. Hazeleger, W. et al. Tales of future weather. Nat. Clim. Chang. 5, 107–113 775–779 (2012). (2015). 82. Lu, J., Chen, G. & Frierson, D. The position of the midlatitude storm track and 112. Overland, J. E., Francis, J. A., Hanna, E. & Wang, M. The recent shift in early Eddy-driven westerlies in aquaplanet AGCMs. J. Atmos. Sci. 67, 3984–4000 summer Arctic atmospheric circulation. Geophys. Res. Lett. 39, L19804 (2010). (2012). 83. Grise, K. & Polvani, L. M. The response of midlatitude jets to increased CO : 113. Pithan, F. et al. Select strengths and biases of models in representing the Arctic distinguishing the roles of sea surface temperature and direct radiative forcing. winter boundary layer over sea ice: the Larcform 1?single column model Geophys. Res. Lett. 41, 6863–6871 (2014). intercomparison. J. Adv. Model. Earth Syst. 8, 764–785 (2016). 84. Butler, A. H., Thompson, D. W. J. & Heikes, R. The steady-state atmospheric 114. Runge, J. et al. Identifying causal gateways and mediators in complex spatio- circulation response to climate change–like thermal forcings in a simple temporal systems. Nat. Commun. 6, 8502 (2015). general circulation model. J. Clim. 23, 3474–3496 (2010). 115. Routson, C. et al. Changing temperature gradients linked to Holocene 85. Newson, R. L. Response of a general circulation model of the atmosphere to moisture trends in the Northern Hemisphere. In American Geophysical Union, removal of the Arctic ice-cap. Nature 241,39–40 (1973). Fall General Assembly 2016 (AGU, 2016). 86. Vallis, G. K., Zurita-Gotor, P., Cairns, C. & Kidston, J. Response of the large- 116. Trouet, V., Babst, F. & Meko, M. Recent enhanced high-summer North scale structure of the atmosphere to global warming. Q. J. R. Meteorol. Soc. Atlantic Jet variability emerges from three-century context. Nat. Commun. 9, 141, 1479–1501 (2015). 180 (2018). 87. Palmer, T. N. Nonlinear dynamics and climate change: Rossby’s legacy. Bull. 117. Hansen, J., Sato, M. & Ruedy, R. Perception of climate change. Proc. Natl Am. Meteorol. Soc. 79, 1411–1423 (1998). Acad. Sci. USA 109, E2415–E2423 (2012). 88. Molnos, S. et al. A network-based detection scheme of the jet stream core. 118. Lobell, D. B., Schlenker, W. & Costa-Roberts, J. Climate trends and global crop Earth Syst. Dyn. 8,75–89 (2017). production since 1980. Science 333, 616–620 (2011). 89. Archer, C. L. & Caldeira, K. Historical trends in the jet streams. Geophys. Res. 119. Kossin, J. P. A global slowdown of tropical-cyclone translation speed. Lett. 35,1–6 (2008). Nature 558, 104–108 (2018). 90. Fu, Q. & Lin, P. Poleward shift of subtropical jets inferred from satellite- 120. Fannin, B. Updated 2011 Texas agricultural drought losses total $7.62 observed lower-stratospheric temperatures. J. Clim. 24, 5597–5603 (2011). billion. AgriLifeTODAY March 21. http://today.agrilife.org/2012/03/21/ 91. Bender, F. A.-M., Ramanathan, V. & Tselioudis, G. Changes in extratropical updated-2011-texas-agricultural-drought-losses-total-7-62-billion/ (2012). storm track cloudiness 1983-2008: observational support for a poleward shift. 121. Rippey, B. R. The U.S. drought of 2012. Wea. Clim. Extremes 10,57–64 Clim. Dyn. 38, 2037–2053 (2011). (2015). 92. Zappa, G., Lucarini, V. & Navarra, A. Baroclinic stationary waves in 122. Zscheischler, J. & Seneviratne, S. I. Dependence of drivers affects risks aquaplanet models. J. Atmos. Sci. 68, 1023–1040 (2011). associated with compound events. Sci. Adv. 3, e1700263 (2017). 93. Wang, S. Y. S. et al. An intensified seasonal transition in the Central U.S. that 123. Leonard, M. et al. A compound event framework for understanding extreme enhances summer drought. J. Geophys. Res. 120, 8804–8816 (2015). impacts. WIREs Clim. Change 5, 113–128 (2014). 94. Lee, M. H., Lee, S., Song, H.-Y. & Ho, C.-H. The recent increase in the 124. Koster, R. D. et al. GLACE Team, Regions of strong coupling between soil occurrence of a boreal summer teleconnection and its relationship with moisture and precipitation. Science 305, 1138–1140 (2004). temperature extremes. J. Clim. 30, 7493–7504 (2017). 125. Koster, R. D., Chang, Y., Wang, H., & Schubert, S. D. Impacts of local soil 95. Wang, S. Y., Davies, R. E. & Gillies, R. R. Identification of extreme moisture anomalies on the atmospheric circulation and on remote surface precipitation threat across midlatitude regions based on short-wave meteorological fields during boreal summer: A comprehensive analysis over circulations. J. Geophys. Res. Atmos. 118, 11059–11074 (2013). North America. J. Clim. 29, 7345–7364 (2016). 96. Masato, G., Hoskins, B. & Woollings, T. Winter and summer Northern 126. Deser, C., Terray, L. & Phillips, A. S. Forced and internal components of hemisphere blocking in CMIP5 models. J. Clim. 26, 7044–7059 winter air temperature trends over North America during the past 50 years: (2013). Mechanisms and implications. J. Clim. 29, 2237–2258 (2016). NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 11 REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 127. Lehner, F., Deser, C. & Terray, L. Toward a new estimate of ‘‘time of 144. Huang, C. S. & Nakamura, N. Local finite-amplitude wave activity as a emergence’’ of anthropogenic warming: insights from dynamical adjustment diagnostic of anomalous weather events. J. Atmos. Sci. 73, 211–229 (2016). and a large initial-condition model ensemble. J. Clim. 30, 7739–7756 (2017). Acknowledgements 128. Merrifield, A., Lehner, F., Xie, S. P. & Deser, C. Removing circulation effects to This work was supported by the German Federal Ministry of Education and Research assess central U.S. land-atmosphere interactions in the CESM Large (BMBF), grant 01LP1611A (G.D.C. and D.C.) and the Netherlands Organisation for Ensemble. Geophys. Res. Lett. 44, 9938–9946 (2017). Scientific Research (NWO), grant 016.Vidi.171011 (D.C.). S.W. is supported by the the 129. Bandyopadhyay, S., Kanji, S. & Wang, L. The impact of rainfall and U.S. Department of Energy Office of Science, Biological and Environmental Research temperature variation on diarrheal prevalence in Sub-Saharan Africa. Appl. program (BER) grant DESC0016605 and the Utah Agricultural Experiment Station, Utah Geogr. 33,63–72 (2012). State University, with partial support from the Central Weather Bureau of Taiwan. 130. Allen, C. D. et al. A global overview of drought and heat-induced tree Further support was provided by the U.S. National Science Foundation, grants PLR- mortality reveals emerging climate change risks for forests. For. Ecol. 1304398 and AGS-1640452 (S.V.). L.W. was supported by NASA grant 80NSSC17K0267, Management 259, 660–684 (2010). NSF grant AGS-1552385, and a grant from the Harvard Global Institute. 131. Flannigan, M. D., Krawchuk, M. A., de Groot, W. J., Wotton, B. M. & Gowman, L. M. Implications of changing climate for global wildland fire. Int. J. Wildland Fire 18, 483–507 (2009). 132. Anyamba, A. et al. Recent weather extremes and impacts on agricultural Author contributions production and vector-borne disease outbreak patterns. PLOS ONE 9, e92538 D.C. initiated this review paper and developed the concept, and all authors (D.C., G.D.C., (2014). S.V., L.W. and S.W.) contributed equally to the writing of the manuscript. 133. Donat, M. G. et al. Global land-based datasets for monitoring climatic extremes. Bull. Am. Meteorol. Soc. 94, 997–1006 (2013). Additional information 134. Kalnay, E. et al. The NCEP/NCAR 40-year reanalysis project. Bull. Am. Competing interests: The authors declare no competing interests. Meteorol. Soc. 77, 437–471 (1996). 135. Hurrell, J. W. Decadal trends in the North Atlantic Oscillation regional Reprints and permission information is available online at http://npg.nature.com/ temperatures and precipitation. Science 269, 676–679 (1995). reprintsandpermissions/ 136. Thompson, D. W. J. & Wallace, J. M. Regional climate impacts of the Northern Hemisphere annular mode. Science 293,85–89 (2001). Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in 137. Ogi, M., Yamazaki, K. & Tachibana, Y. The summertime annular mode in the published maps and institutional affiliations. Northern Hemisphere and its linkage to the winter mode. J. Geophys. Res. D. Atmos. 109,1–15 (2004). 138. Folland, C. K. et al. The summer North Atlantic oscillation: past, present, and future. J. Clim. 22, 1082–1103 (2009). Open Access This article is licensed under a Creative Commons 139. Pfahl, S., Schwierz, C., Croci-Maspoli, M., Grams, C. M. & Wernli, H. Attribution 4.0 International License, which permits use, sharing, Importance of latent heat release in ascending air streams for atmospheric adaptation, distribution and reproduction in any medium or format, as long as you give blocking. Nat. Geosci. 8, 610–614 (2015). appropriate credit to the original author(s) and the source, provide a link to the Creative 140. Rohli, R. V., Wrona, K. M. & Mchugh, M. J. January northern hemisphere Commons license, and indicate if changes were made. The images or other third party circumpolar vortex variability and its relationship with hemispheric temperature material in this article are included in the article’s Creative Commons license, unless and regional teleconnections. Int. J. Climatol. 25,1421–1436 (2005). indicated otherwise in a credit line to the material. If material is not included in the 141. Barnes, E. A. Revisiting the evidence linking Arctic amplification to extreme article’s Creative Commons license and your intended use is not permitted by statutory weather in midlatitudes. Geophys. Res. Lett. 40, 4734–4739 (2013). regulation or exceeds the permitted use, you will need to obtain permission directly from 142. Nakamura, N. Two-dimensional mixing, edge formation, and permeability the copyright holder. To view a copy of this license, visit http://creativecommons.org/ diagnosed in an area coordinate. J. Atmos. Sci. 53, 1524–1537 licenses/by/4.0/. (1996). 143. Nakamura, N. & Zhu, D. Finite-amplitude wave activity and diffusive flux of potential vorticity in Eddy–Mean flow interaction. J. Atmos. Sci. 67, © The Author(s) 2018 2701–2716 (2010). 12 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

The influence of Arctic amplification on mid-latitude summer circulation

Loading next page...
 
/lp/springer-journals/the-influence-of-arctic-amplification-on-mid-latitude-summer-2Vn60k75GQ

References (200)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2041-1723
DOI
10.1038/s41467-018-05256-8
Publisher site
See Article on Publisher Site

Abstract

REVIEW ARTICLE DOI: 10.1038/s41467-018-05256-8 OPEN The influence of Arctic amplification on mid-latitude summer circulation 1,2 1,2 3 4 5 D. Coumou , G. Di Capua , S. Vavrus , L. Wang & S. Wang Accelerated warming in the Arctic, as compared to the rest of the globe, might have profound impacts on mid-latitude weather. Most studies analyzing Arctic links to mid-latitude weather focused on winter, yet recent summers have seen strong reductions in sea-ice extent and snow cover, a weakened equator-to-pole thermal gradient and associated weakening of the mid-latitude circulation. We review the scientific evidence behind three leading hypotheses on the influence of Arctic changes on mid-latitude summer weather: Weakened storm tracks, shifted jet streams, and amplified quasi-stationary waves. We show that interactions between Arctic teleconnections and other remote and regional feedback processes could lead to more persistent hot-dry extremes in the mid-latitudes. The exact nature of these non- linear interactions is not well quantified but they provide potential high-impact risks for society. he observed increases in the frequency and intensity of extreme heat and heavy rainfall events since the late 1980s, especially in mid-latitude regions, have been linked to 1–3 Tanthropogenic global warming . Scientists are generally confident in the thermo- 4,5 dynamic drivers of these changes but are less so in dynamic aspects . Another pronounced signal of anthropogenic global warming is the rapidly increasing near-surface temperatures in the Arctic at a pace two to four times faster than the rest of the globe, known as Arctic amplification (AA) . The extent to which AA affects the mid-latitude circulation and possibly contribute to the observed increases in weather extremes has been a subject of active debate . Most studies analyzing the role of AA on mid-latitude weather have focused on the winter season and the linkage with cold spells. The stronger jet stream, the presence of the stratospheric polar vortex, and the post-1990s increase in abnormally cold winters over central Eurasia have 6,8,9 drawn a lot of attention to the winter season . The increased heat stored in the Arctic Ocean owing to sea-ice loss is released into the atmosphere in early winter. The associated expansion of the near-surface air increases Arctic geopotential heights and can affect the circumglobal cir- culation directly as well as via feedbacks between the troposphere and stratosphere involving the 6,10–14 stratospheric polar vortex . Even though the exact pathways through which the Arctic influences the mid-latitude winter circulation are debated, a scientific consensus is emerging that 7,15,16 AA has at least some influence on winter weather . Links between AA and summer circulation have received far less scientific attention, despite the potential for synergistic effects that might favor high-impact extremes. In summer, thermodynamic 1 2 Department of Water & Climate Risk, Institute for Environmental Studies, VU Amsterdam, Amsterdam 1087HV, Netherlands. Department of Earth System Analyses, Potsdam Institute for Climate Impact Research, Potsdam 14473, Germany. Nelson Institute Center for Climatic Research, University of Wisconsin-Madison, Madison 53706 WI, USA. Department of Earth and Planetary Sciences, Harvard University, Cambridge 02138 MA, USA. Department of Plants, Soils and Climate, Utah State University, Logan 84322 UT, USA. Correspondence and requests for materials should be addressed to D.C. (email: coumou@pik-potsdam.de) NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 1 1234567890():,; REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 BOX 1 | Recent mid-latitude summer weather extremes and their impacts Many recent high-impact summer heatwaves occurred in the far-tail of the distribution and are difficult to explain by the direct radiative warming effect 18–21 of greenhouse gas forcing alone . In 2010, Russia saw 33 consecutive hot-and-dry days (with temperatures above 30 °C), resulting in an estimated 55,000 heat-related deaths, more than 500 wildfires near Moscow and grain-harvest losses of 30% . A quantitative global analyses showed that the 2010 event was the most-severe heatwave ever recorded worldwide, based on a heatwave index that can be used across different regions . Intriguingly, all record-setting heatwaves based on this index occurred in the mid-latitudes, indicating that here heatwaves are becoming more intense at a pace that exceeds the global mean . Extreme summer heat in the Northern Hemisphere mid-latitudes now far exceeds historical frequencies in the twentieth century . Over the last decade, Europe has seen an exceptionally rapid increase in the chance of extremely hot summers similar to the 2003 21 4,20 extreme . Other notable high-impact and record-breaking droughts and heatwaves occurred in the USA in 2011 and 2012 , leading to billions of 120,121 dollars in agricultural losses . One reason why these extremes cause so much damage is that temperature and precipitation during summer are anti-correlated virtually everywhere across extratropical land . Consequently, extremes of heat and dryness often coincide to produce compound extremes that exert disproportionately 123 124 large societal impacts . Certain mid-latitude regions identified as hotspots of tight atmosphere–land coupling, such as central North America , are especially prone to concurrent heatwaves and droughts. Such events are promoted by positive thermodynamic feedbacks that favor depleted soil moisture and enhanced sensible surface heating as the land warms, but they are also strongly regulated by atmospheric dynamics that initiate and sustain anomalous heating and drying . These kinds of summertime compound events are therefore highly relevant for the dynamical changes described in this study, and recent studies have developed 126–128 innovative techniques to separate the contributions from dynamics and thermodynamics . In particular, the projected trend toward a weaker and poleward-shifted jet stream is consistent with projections of a significantly increased risk of compound hot-dry extremes across much of the Northern Hemisphere this century . This type of climate change would likely exacerbate the separate impacts of extreme heat and dryness, based on the 129 130 131 documented stresses that compound heatwaves and droughts exert in causing disease , vegetation mortality , wildfires , and agricultural losses . and dynamic drivers of extreme weather could act in the same when near-surface air temperatures drop below sea-surface direction, leading to tail risks . For instance, any increased fre- temperatures, this excessive heat is released into the 6,10–13 quency in circulation regimes conducive to persistent heat extremes atmosphere . The additional heat inflates the lower tropo- would act on top of the thermodynamically driven increase in heat, sphere over the Arctic Ocean and nearby continents, and increase creating possibilities for very-extreme heatwaves. Many recent high- geopotential heights, which could affect circulation patterns fur- impact summer heatwaves indeed occurred in that far-tail of the ther south. 6,14 distribution and cannot be explained by the direct thermodynamic The observed increase in Arctic geopotential heights might 18–21 effect of greenhouse gas forcing alone (Box 1) .Suchextreme reduce the poleward pressure gradient in the troposphere and heatwaves have been found to increase and intensify across most therefore weaken the storm tracks and westerly jet. However, this regions but more so in the mid-latitudes than over the rest of the notion based on thermal-wind balance and baroclinicity provides globe . Consistent with the increase in heatwaves, the hot tail of little explanation for the recent changes in winter circulation. In summer temperature distribution has been warming faster than the winter, the near-surface warming in the Arctic has been pro- median and the cold tail. Figure 1 shows the warming trends in the nounced but confined to high latitudes only, i.e., north of 70° N . 95th percentile (hot tail), 50th percentile (median), and 5th per- Within the mid-latitudes (i.e., 30° N–60° N), neither the poleward centile (cold tail) of daily summer temperatures. Clearly, over most temperature gradient nor the zonal-mean jet or storm track have mid-latitude regions, in particular over Eurasia but less so in the US, seen any significant changes in winter . Future climate model the hot tail has been warming faster than the cold tail and thus projections under high-emission scenarios show that both chan- 22 27,28 29–31 temperature variability in summer has increased . This increased ges in the tropics and in the Arctic can influence the variability indicates that more complex processes beyond simple strength and position of the mid-latitude winter circulation. radiative greenhouse gas forcing are important in driving heat Enhanced warming projected in the tropical tropopause region extremes (Box 1). This is supported by recent studies that indicate (due to enhanced deep convection and latent heating) acts to that summer weather has become more persistent in several regions increase the upper-level poleward temperature gradient, which 23–25 28 in the mid-latitudes . In summer, the hot tail of the distribution strengthens the mid-latitude westerlies . This has become known is associated with persistent, blocking weather systems, and an as the tug-of-war (Fig. 2), whereby tropical changes tend to increase in their persistence leads to more extreme temperatures. strengthen mid-latitude circulation and lead to a poleward Here, we review recent studies analyzing possible links between migration, whereas AA has the opposite effect . AA, mid-latitude summer weather and extreme events, in particular Many recent winters were characterized by extremely warm persistent hot-dry extremes. We start by giving a brief synopsis of temperatures in the Arctic and anomalously cold conditions the different Arctic mechanisms proposed for winter. Next, we further south, especially over Eurasia. In fact, large areas over address the seasonal differences in the mid-latitude circulations, in central Eurasia have been cooling since 1990 . Possible dyna- the influence of regional and far-away drivers, and in detected mical mechanisms behind this warm-Arctic cold-continent changes in the Arctic. We focus on three possible dynamical pattern involving sea-ice loss consist of a direct tropospheric pathways that are most relevant to summer and summarize the pathway and pathways involving the stratospheric polar theoretical, empirical, and modeling evidence for each of them. We vortex (Fig. 2) . In the latter hypothesis, increased geopotential discuss the confidence and uncertainties associated with these heights over high-latitude regions can cause a pronounced dynamical pathways, identify knowledge gaps and key societal risks, upward wave propagation into the stratosphere which can and provide a roadmap for future research. weaken the stratospheric polar vortex and, in extreme cases, trigger sudden stratospheric warming events. A weak polar vortex can propagate downwards into the troposphere causing a Arctic amplification and mid-latitude winter circulation negative Arctic oscillation (AO) that is conducive to cold spells Due to declining sea-ice, the Arctic Ocean absorbs more in Eurasia and Siberia . incoming solar radiation from spring to autumn. By early winter, 2 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications U250 hPa U250 hPa v250 hPa v250 hPa T2m T2m U30 hPa NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE a d 95% 95 – 50% qqq q q q q q q q q q q q q q q q qq q q q q q q q q q qq q q q q qq q q q q q q q q q qq q q q q q q q q q q q qq q q q q q q q q qq q q q qq q q q q q q q q qq q q q qq qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q qqq q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q b e 50% 95 – 5% q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qqq q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q qq q qq q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q qq q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q c f 5% 50 – 5% q q qqq q qqq q q q q q q q q q q q qq q q q q q q q q q q q q q q q qq q q q qq q q q q q q q q qq q q q q q qq q q q q q q q qq q q q qq q q q q qq q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q q qq q q q −1.5 −1.2 −0.9 −0.6 −0.3 0 0.3 0.6 0.9 1.2 1.5 °C/decade Fig. 1 Summer trends in surface temperature over 1980–2011. a 95th, b 50th, and c 5th quantile of the HadGHCND gridded daily dataset; differences in the trends of different quantiles, plotted in d–f,reflect changes in the width of the distribution. Over most mid-latitude regions, especially over Eurasia, the width of the distribution has broadened and thus variability has increased. (Figure created using R statistical software) Winter Summer ab Monthly long term mean wind speed (m/s) January 20.0 30.0 40.0 50.0 60.0 70.0 July 12.0 16.0 20.0 24.0 28.0 32.0 Stronger jets & a polar vortex Weaker subtropical jet & double jets cd Monthly long term mean v wind (m/s) January –12.5 –7.5 –2.5 2.5 7.5 12.5 July –7.5 –4.5 –1.5 1.5 4.5 7.5 January T2m 230.0 244.0 258.0 272.0 286.0 300.0 July T2m 265.0 272.0 279.0 286.0 293.0 300.0 Longer stationary waves Shorter stationary waves Weaker land–sea heat contrast Stronger land–sea heat contrast Fig. 2 Schematic figure illustrating the main seasonal differences in upper tropospheric circulation between winter (January) and summer (July). Panels a and b show 250-hPa wind speed (green-to-blue shading) illustrating the jet streams with black arrow lines that follow the zone of maximum wind speed. The wintertime stratospheric polar vortex is outlined with the thick green line following the 30-hPa maximum wind speed. Panels c and d show the 250 hPa meridional wind speed (dark gray-to-dark red shading) depicting the stationary wave features associated with the jet streams. White arrows are addedto illustrate wind direction. Basic differences in the summer circulation features, as compared to winter, include shorter stationary waves, more northerly subtropical jet, absence of stratospheric polar vortex and an Arctic front jet forming double jets. Data are 1970–2000 climatology of NCEP Reanalysis (downloadable: https://www.esrl.noaa.gov/psd/data/gridded/data.ncep.reanalysis.html). (Figure created using Panoply and Apple’s Keynote software) NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 3 REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 There is relatively high confidence that this stratospheric and severe weather outbreaks (e.g., flooding) due to their 56,57 pathway is a real phenomenon and has a role in the warm- longer lifetime than synoptic disturbances . Arctic cold-continent pattern, as it is supported by The climatological jets in summer have less northward tilt multiple lines of evidence: From empirical analyses, to causal compared to winter and therefore waveguides are also oriented 6,15,35,36 discovery algorithms, and climate model simulations . west-to-east and have the potential to become circumglobal Still, its relative importance compared to other pathways, (Fig. 2). Also, the narrower mean jet in summer favors the tropical influences and atmosphere internal variability remains waveguide effect with wave trains orienting zonally along the jet 7 38,56,58,59 unclear . stream waveguide . However, the zonal orientation of the climatological jet does not mean that the overall flow is less wavy in summer compared to winter: The total waviness in geopo- Summer circulation tential height fields on sub-synoptic to sub-seasonal time scales is 60–62 Compared to winter, summer circulation in the mid-latitude is as pronounced in summer as in winter, if not more .In weaker, more barotropic and the climatological jets are more winter, Rossby waves are typically oriented along a meridional zonally oriented, which promotes the formation of circumglobal path arcing from the tropics into the mid-latitudes, or from the wave trains (CGWT, see Box 2). It is less influenced by variability mid-latitudes into the tropics (see Fig. 3 of Hoskins and Wool- in tropical sea-surface temperatures (SST), and more sensitive to lings ). Due to the zonal orientation in summer, any local land-atmosphere feedbacks involving soil moisture or snow heating anomaly can generate a sequence of waves of similar cover. wavelength downstream of the jet, forming a stagnant wave In summer, the reduced pole-to-equator temperature gradient packet that affects weather conditions far away. Recent research (as compared to winter) leads to weaker and more-narrow upper- has shown that when synoptic-scale waves (wavenumbers 6–8) level westerlies, and the stratospheric polar vortex is absent all are trapped in a (near) circumglobal waveguide, wave-resonance 19,64–66 together (Fig. 2). The role of the stratosphere in influencing can greatly increase their amplitude . The wave-resonance boreal summer weather is therefore considered non-existent . mechanism can lead to highly persistent and anomalous weather Moreover, variations in tropical SST have less influence on mid- conditions around the hemisphere and studies have linked it to latitude circulation in summer compared to winter (e.g., several recent high-impact summer extremes, including heat- 38,39 57,66 refs. ). The position of abnormally warm SST in the tropics waves and floods . Though there is a solid theoretical basis determines where the strongest deep convection takes place underlying wave-resonances, their exact significance in the real- associated with shifts in the Walker circulation. The upper-level world in causing extreme weather events is debated . latent heat release during deep convection can trigger long Rossby Finally, the Arctic frontal zone that develops around 70° N in waves that propagate poleward and influence mid-latitude summer is likely to be affected by AA. The warm, snow-free weather . This mechanism is less important in summer than land surface and the cold Arctic Ocean create a strong thermal winter for two reasons: First, the El Niño-Southern Oscillation contrast along the Arctic seaboard around 70° N, generating (ENSO), the dominant mode of variability in tropical SST, tends strong westerlies here. These sub-polar westerlies, together with to peak in boreal winter and is much weaker during boreal the pronounced sub-tropical jet, form the distinct summer feature summer . Second, the prevailing easterly winds in the tropics in of double jets (Fig. 2). A double-jet regime is characterized by a summer limit the ability of Rossby waves to propagate pole- very confined sub-tropical jet with sharp edges wherein wind ward . This is not to say that the tropics cannot influence speeds change rapidly with latitude. Such sharp sub-tropical jets summer mid-latitude weather. Among other things, ENSO gives and thus double jets favor waveguide are effective waveguides 42,43 69 some predictive seasonal forecast skill in summer and the formation and wave-resonance events . Interactions of the two varying location and intensity of monsoon systems, notably over jets can produce high-amplitude atmospheric waves, creating the 66,70 South and East Asia, can affect the mid-latitude summer circu- deepening of troughs and stagnation of ridges . 44–46 lation . Given these specific characteristics of the summer circulations, The state of the cryosphere, in terms of sea-ice and snow cover, several mechanisms have been proposed that link AA with from late winter to early summer can influence the strength and summer mid-latitude weather patterns. These are grouped into latitude of the summer time jet . Boreal snow cover during weakening of the storm tracks, shift in the latitudinal position of spring and summer has shrunk dramatically in recent years, even the mid-latitude jet, and amplification of circumglobal wave 48,49 faster than the decline of Arctic sea-ice extent . Without snow, trains (Fig. 3). the surface albedo is lower and thus the land regions absorb more incoming solar radiation. Furthermore, declining snow cover in spring has a delayed drying effect on the soils by mid-summer, Influence of the Arctic on summer circulation favoring enhanced temperatures due to suppressed evaporative Weakening storm tracks. Theoretical, observational and mod- 50,51 cooling . These thermodynamic processes in conjunction with eling evidence supports the hypothesis that summer storm tracks 26,71,72 reductions in early season snow cover can affect regional to weaken with enhanced Arctic warming . The theoretical 52,53 hemispheric circulation . basis underlying AA and resultant weakening of the mid-latitude The waveguide effect, i.e., the trapping and focusing effects of storm track is straightforward: The thermal-wind balance relates the seasonal jet streams on low-frequency tropospheric waves, vertical shear in the westerly flow to the magnitude of the pole- has an important role in how a changing mid-latitude circu- ward temperature gradient. In the lower troposphere, a reduction 39,54 lation might promote stagnant weather patterns .Wave- in the temperature gradient equates to a similar reduction in the guides produce zonally oriented chains of perturbations that shear, weakening the thermally driven jet and reducing the low- fluctuate at relatively low frequency ranging from weeks to level baroclinicity . A reduced low-level baroclinicity implies less months, creating a teleconnection pattern. Early research found or weaker synoptic-scale cyclogenesis and thus leads to overall that atmospheric disturbances near the jet core are refracted weakening of the storm tracks. Note, that the thermal-wind toward the core, meaning that the jet acts as a waveguide .The balance does not give a direction of causality per se: The causality energy of the trapped disturbances does not disperse strongly could be the other way around, whereby a change in mid-latitude and therefore can propagate much further and possibly circulation alters the poleward heat transport giving rise to more become circumglobal. Such CGWT can generate heatwaves rapid warming in the Arctic . 4 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications P NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE In past climate In warming climate AA Weakening of storm tracks Storm systems Relatively warm Arctic Storm track Relatively cool Arctic Equatorward shift of polar jet AA vs. poleward shift of subtropical jet Amplification of quasi- AA stationary short-wave trains Arctic circle warming over land Fig. 3 Schematic representation of proposed dynamical mechanisms in summer. a Weakening of storm tracks, b latitudinal-shift in jet positions, and c amplification of quasi-stationary waves (Figure created using Apple’s Keynote software) Empirical evidence based on multiple datasets shows that over the satellite-covered period (i.e., since 1979), the mid-latitude summer circulation has indeed weakened in conjunction with a reduction in the poleward temperature gradient in the lower troposphere. This weakening has been detected in the westerly jet (following the thermal-wind balance), the total kinetic energy of –5 synoptic storm systems (by about 15%) and the number of strong 26,71,74 cyclones . Similarly, strong Arctic sea-ice melting years are 75 –10 characterized by a weakened circulation . While, the satellite era is most reliable when analyzing wind field characteristics, its limited timespan compromises long-term trend analyses. Natural –15 variability on multi-decadal time scales, either due to changes in SSTs or from internal atmospheric variability, are thus likely to –20 have a role in the observed trends. There is modeling evidence indicating that these observed trends are at least partly attributable to AA. CMIP5 coupled –25 model simulations of the twentieth century show that the observed changes in the zonal-mean temperature gradient in Slope = 1.4 –30 summer (characterized by AA and enhanced high-latitude land p val = 3.5e–05 warming) are likely attributable to anthropogenic forcing ("likely" according to IPCC lexicon) . Idealized modeling experiments –10 –5 0 5 support storm track weakening when sea-ice is reduced but also Δ U (%) indicate that sea-ice changes by itself can explain only part of the Fig. 4 Observed and projected changes in the mid-latitude Northern observed weakening . Modeling studies indicate that the effects Hemisphere summer storm tracks and westerlies. The percentage of historic sea-ice reductions can explain up to one-third of the change in summer storm tracks (vertical axis) and westerlies (horizontal magnitude of the observed anomalies, with an additional role for 72,78 axis) in future (2081–2100, under scenario RCP8.5) relative to changes in SSTs . Thus, other factors including natural 1981–2000 for individual CMIP5 climate models is shown, and their variability likely had a role in the recently observed summer linear fit (solid black line). Observed changes based on ERA-Interim data circulation changes, but a substantial share of it is likely are given for the 1979–2013 period. Taken from (Coumou et al. .) attributable to AA. NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 5 Δ EKE (%) REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 6 2 –1 10 m s 1.6 1.2 0.8 0.4 –0.4 –0.8 –1.2 –1.6 –2 Fig. 5 Enhanced circumglobal wave train embedded in the summer jet. Linear trends from 1979 to 2010 in the July 250 hPa stream function in the short- wave regime (blue-red shading) computed with the long wavenumbers (1–4) removed. The change in the short waves is embedded in the climatological July-mean 250-hPa wind speed depicting the jet stream (black contour lines). Adapted from (Wang et al. .) 90,91 For future high-emission scenarios, models robustly project observations of clouds and is most significant in winter. storm track weakening, supporting the hypothesis that AA is This is thus largely in agreement with model projections pointing associated with weakened summer storm tracks (see Fig. 4) at the important role of the tropics in shifts in the jet position. 26,71,79,80 . The changes at the end of the century in the high- emission scenario are comparable to the observed changes over Amplification of wave trains. Limited evidence from theory, the past decades . This suggests that either the models under- observations and some model simulations suggests that AA may estimate the future changes (models also underestimate historic amplify synoptic-scale, quasi-stationary waves embedded in the changes in the Arctic itself) or that a substantial part of the summer jet (Fig. 5). Theory of the dynamics of a dry atmosphere observed trend is associated with multi-decadal natural varia- suggests that a lower troposphere diabatic heating source in the bility . For the North American sector, the amount of AA in mid-latitudes will have a larger stationary wave response (in models by the end of century is negatively correlated with the terms of a meridional stream function displacement) when the changes in jet speed and wave phase speed in summer . Models background baroclinicity and zonal winds are reduced, as a direct robustly project a weakening of summer storm tracks by the end response of AA . In a more complex atmosphere of an aqua- of the century (Fig. 4), but this is not the case for the upper-level planet model (i.e., an Earth covered by water only), quasi- jet: There is large inter-model spread with some models stationary synoptic-scale wave trains are enhanced when the projecting a strengthening and some a weakening of the upper- meridional temperature gradient is reduced and the westerly 26,28 level jet (Fig. 4) . winds weaken . Thus, as the background flow becomes weaker, the same heating source in the lower troposphere can trigger a Shift in jet-position. Understanding jet shifts in the Northern stronger stationary wave response especially for synoptic-scale Hemisphere is a challenging task due to several competing pro- waves. The increased moisture content in a warmer atmosphere, cesses. Theoretically, the change in jet stream position can be and the tendency for increased latent heat release in the tropics divided into a relatively small direct radiative-induced shift and a and over warm ocean currents in higher latitudes, can provide larger indirect SST-mediated effect. The indirect effect comes further heating to perturb more or stronger CGWTs. from mid-latitude dynamical feedbacks, involving enhanced Some observational evidence suggests that the quasi-stationary irreversible mixing due to wave breaking of high-frequency component of mid-latitude summer circulation has become transient eddies , and can explain most of the expected jet wavier since 1979, in particular over the North American 83 60,62,93,94 shifts . When considered in isolation, AA should theoretically sector . Figure 5 plots linear trends over 1979–2010 in cause a southward shift in the mid-latitude jet stream. the short-wave regime showing enhanced CGWTs over both the Idealized dry atmospheric model simulations indeed indicate a North American and Eurasian sectors. For the American sector, southward shifted jet coming from AA by itself . This is also this is further supported by detected increases in waviness 60,62 confirmed by more complex models with reductions in sea-ice metrics (see Box 3). Using three different climate models, the 31,72,78,85 imposed . Despite the process linking AA to a more summer time amplification of quasi-stationary short waves over equatorward jet, the zonal-mean jet streams are projected to the American sector appears to be attributable to greenhouse gas migrate poleward by about one degree by the end of the twenty- forcing . As the CGWT is linked to the summer North Atlantic 27,86 39 first century under a high-emission scenario . Thus in the long Oscillation (SNAO, Box 2) , an AA-like circulation anomaly run, i.e., at the end-of-century, the tropics likely dominate the (i.e., negative phase of the SNAO) could modulate the CGWT tug-of-war, at least in models. While in winter both the Atlantic and vice versa. However, the summer CGWT can also be and the Pacific jets are projected to migrate poleward, in summer triggered by heating sources associated with the Indian 28 44 this shift is seen only for the North Atlantic jet . AA may monsoon . therefore exert a stronger opposing influence to the expected The few modeling studies with historically observed low sea-ice poleward shift of the Pacific jet in summer. Different state-of-the- concentration show a stationary wave-train response in summer 78,95 art climate models employ different simplifying parameteriza- emerging from AA . Still, future model projections of mid- tions (e.g., for clouds) and those can, in a complex, non-linear latitude quasi-stationary short-wave patterns are generally system, lead to very different outcomes . The inter-model spread inconsistent with recent observations. Future CMIP5 projections of the poleward shift therefore tends to be larger than the signal under high-emission scenarios show an overall decrease in 27,46,86 96 itself . blocking both for winter and summer . The reasons behind these Observed jet shifts in the Northern Hemisphere are generally divergent findings are not well understood and may result from small compared to those in the Southern Hemisphere, but still competing effects from the tropical monsoons , changes in land- 88 46 generally indicate a poleward migration . For the Northern sea thermal contrast , model biases in representing summer time Hemisphere, some evidence of a poleward shift of jet streams Rossby waves and the use of different diagnostics to quantify 88,89 has been identified in reanalysis products and satellite waviness (see Box 3). 6 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE BOX 2 | Atmospheric teleconnections in summer Remote climate effects, known as atmospheric teleconnections, consist of three main types: (i) regional long-wave pattern confined in a longitudinal sector of the globe, like the Pacific-North America pattern and the North Atlantic Oscillation ; (ii) a hemisphere-wide pattern with a prominent zonal- 70,136,137 mean component like the Arctic Oscillation or the Annular Modes ; and (iii) the trapping and focusing effects of the seasonal jet streams on 39 54 low-frequency tropospheric waves , known as the waveguide effect . The figure below shows schematic representations of these three types of teleconnection patterns. The summer North Atlantic Oscillation (SNAO) can be regarded as the counterpart of the more robust winter NAO. The centers of action of the SNAO exhibit a more northerly location, have a smaller geographical extent and a weaker dipole pattern compared to the winter NAO. Panel a schematically shows the surface pressure patterns of the positive SNAO phase, i.e., when pressure differences are strong, together with the position of the warmer/ drier and colder/wetter regions. Generally, the negative phase shows a reverse pattern. Like the winter NAO, the impact of the SNAO on climate extremes such as heavy rainfall and flooding is profound, especially for Europe. Post-2007 summers have seen increasingly robust negative SNAO associated with a persistent anticyclonic anomaly over Greenland and a cyclonic anomaly over Northwest Europe. This pattern caused rapid melting of the Greenland ice sheet and brought unusually wet summers to Northwest Europe, including the massive flooding of U.K. in summer 2012. Future projections of climate models suggest an increasingly positive SNAO in warmer climates . The Northern Hemisphere Annular Mode (NAM) is an internally driven atmospheric mode maintained by both stationary and transient waves. The NAM is defined as the first EOF of monthly 500 hPa geopotential height fields and coincides with the definition of the Arctic Oscillation during the winter months . Winter and summer NAM patterns present both a diverse geopotential height field, mean meridional circulation and eddy structure. The summer NAM has a smaller latitudinal extent and has a stronger link to surface air temperature over Eurasia. The positive phase of the summer NAM is associated with negative geopotential height anomalies over Greenland and the Arctic Ocean and an annual band of positive anomalies comprised between 40° and 60° N and particularly extended over Eurasia, as schematically shown in panel b. When the summer NAM is strongly positive, the storm tracks follow the Arctic front . Anomalously positive summer NAM phases are associated with double jets favoring blocking between the polar and the subtropical jets. During a positive summer NAM phase, surface temperatures over Eurasia show a dipole pattern with warmer conditions over Europe and colder conditions over East Asia . The energy of waves trapped in a waveguide is not dispersed as broadly as in teleconnections of type (i) and therefore it can propagate farther before being dissipated. In summer, when an efficiently trapping waveguide becomes (almost) circumglobal, then resonant interactions between free and forced waves (typically of synoptic scale, wavenumbers 6–8) might lead to wave-amplification and persistent, high-amplitude waves. Waveguide formation is favored during double-jet regimes and thus linked to strongly positive NAM phases. Panel c shows a schematic representation of a wave- resonance event with an amplified wave 7. The orange and green areas represent regions of positive and negative upper-level meridional winds. Such events are associated with alternating hot-dry and cold-wet conditions, following the ridges and troughs. Such a situation is prone to blocking weather systems and deepened troughs, and their relative longevity is key to making severe weather extremes . Wave-resonance periods have therefore been 57,64 linked to both persistent heatwaves and severe flooding events . Finally, blocking itself is also distinctly different in summer compared to winter. Upstream latent heat release has been identified as an important contributor to persistent blocking and this mechanism is especially important in summer . a c SNAO well above zero Summer NAM positive Wave-resonance Colder/wetter Storm tracks Warmer/drier Figure Box 2: Panel a shows a schematic of the positive SNAO indicating the anomalously low (blue) and high (red) sea level pressure regions together with cold/wet and dry/warm regions. Panel b shows a schematic of the 500 hPa geopotential height configuration during the positive phase of the NAM. Rainy clouds mark the position of the storm tracks during a strongly positive NAM phase. Panel c shows a schematic of an amplified wave 7. Green and orange regions show the position of pronounced northward and southward wind anomalies. (Figure created using Python, GIMP, Powerpoint and Inkscape software) Double jets reduction of snow cover. Thus, while the overall equator-to-pole Basic theory suggests that double-jet flow regimes are to become temperature gradient reduces, the thermal gradient actually slightly more common with pronounced AA . Such flow regimes increases at the land-ocean boundary around the Arctic are characterized by sharper sub-tropical jets (i.e., a strong mer- circle. This situation favors the formation of the Arctic front jet at 68,69 idional wind shear) which can act as waveguides . Moreover, ~70° N in addition to the sub-tropical jet which is normally in summer, recent AA is characterized by enhanced land present. Such double-jet regimes have become more frequent in warming over the high latitudes (~70° N) and much less warming recent years due to high-latitude land warming, something which over the nearby Arctic Ocean, in stark contrast to winter warming is partly attributable to anthropogenic greenhouse gas forcing . patterns. This enhanced high-latitude land warming is likely Double jets favor waveguide formation and wave-resonance but related to a combination of the smaller heat capacity of land the evidence for an increase in frequency or persistence of 66,69 compared to ocean, as well as late-spring to early-summer waveguides is limited . NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 7 REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 BOX 3: | Measuring mid-latitude waviness Although the concept of wavy versus zonal circulation patterns is straightforward, quantifying these opposite states has proven challenging. Approaches can be roughly separated into geometric and dynamic methods. The former focuses on the geometry of the circulation to characterize the 61,62,140,141 departure of the flow from zonality in terms of wave amplitude, sinuosity, or circularity . These metrics have the advantage of being intuitive and readily visualized from geopotential height contours, but they have been criticized for lacking a firm physical basis. The existence of wave trains usually lasts 2–4 weeks and they do not necessarily have a preferred phase position . Averaging over time may thus cancel out any signal. Further, 58,64,65 often the diagnostic approach to detect extratropical jet’s waviness involves Fourier decomposition of zonal wavenumbers which could lead to spurious wave signals. In the extreme case: Fourier decomposition of a delta function will create amplitudes in a range of zonal wavenumbers even if the circulation feature is extremely local, i.e., it does not involve a wave “train” . Likewise, a zonal wavenumber 5 along 40° N (subtropical jet) or 65° N (polar jet) depicts two very different wavelengths. Dynamically based waviness metrics such as effective diffusivity of potential vorticity and finite- 143,144 amplitude wave activity are derived from first-order mass and circulation conservation principles and satisfy exact budget closure equations. These measures provide a theoretical basis for quantitatively relating changes in zonal wind speed to accompanying changes in wave amplitude and eddy fluxes, which is straightforwardly verifiable under idealized conditions. Such approaches are being applied in climatological studies of circulation trends and extreme weather events related to amplified flow patterns but their derivation is more technical and their application is more involved than geometric methods. Combined effects The Northern Hemisphere summer circulation is connected to The described dynamical mechanisms do not operate in isolation the tropics primarily via two-way interactions with monsoon but instead interact with each other, with other teleconnections, systems. Anomalously low snow cover over Eurasia during late and regionally with land-atmosphere feedbacks. The exact nature winter to early-spring can increases the land-ocean temperature of these interactions is still speculative at this stage but positive gradient due to snow-albedo and soil moisture effects which feedbacks are certainly possible which would lead to tail risks. strengthens the Indian summer monsoon rainfall . On sub- Weakened storm tracks favor the buildup of hot and dry seasonal time scales, extreme monsoonal rainfall over eastern or conditions over the continents. This can strengthen the land- western Himalayan foothills is often linked to southward ocean thermal contrast which, combined with the enhanced wave intruding mid-latitude troughs, possibly part of the CGWT . response to thermal forcing when the background flow is weak, The monsoon is also a source of diabatic heating that can gen- could lead to amplified quasi-stationary waves. The projected erate the CGWT . Thus, wave trains originating from higher weakening of storm tracks and westerlies can interact with soil latitudes can modulate the intensity of the monsoon, and the moisture and snow cover changes, potentially regionally exacer- monsoon’s strength in turn reinforces the downstream propaga- bating zonal-mean circulation changes. Enhanced terrestrial tion of the wave train . The projected increase in Indian summer heating in mid- to high-latitudes associated with reductions in monsoon rainfall is thus expected to strengthen the CGWT. snow cover can promote anomalous ridging in the upper tropo- sphere, as suggested by both observational and modeling Robust evidence and knowledge gaps 53,97,98 studies . The emergence of such positive geopotential height Several arguments support the AA influence on summer circu- anomalies in mid-to-high latitudes induces hot and dry air lation as compared to winter. First, in the mid-latitudes, the through subsidence and easterly wind anomalies of continental equator-to-pole near-surface temperature gradient has seen a origin . Likewise, drying soils in subtropical regions like the pronounced reduction in summer, but not in winter . Second, Mediterranean (a robust projection in future climate) can favor late-spring to early-summer snow cover extent, which influences 48,109 the formation of a heat low, i.e., low pressures at the surface due summer flow regimes, has dramatically declined ; and finally, to upflow associated with intense surface heating . A Medi- summer circulation is less affected by tropical ENSO forcing, and 17,38,44 terranean near-surface heat low brings easterly winds to central thus potentially more sensitive to Arctic warming . and western Europe, obstructing the normal westerlies . Wea- Reviewing the literature, we conclude that there is robust evi- kened or diverted westerlies can reduce rainfall leading to further dence that AA causes a weakening of storm tracks in summer. soil drying. Due to such feedback mechanisms between soil The mechanism is straightforward: Cyclone genesis is directly moisture, snow cover changes, and continental-scale circulation, related to the lower troposphere temperature gradient which summer weather in western Europe and interior North America weakens with AA. Further support comes from observed trends 60,100,101 is likely to become more continental as seen in models and historic and future climate modeling experiments. Still, 24,25,102 and possibly also in observations . The effectiveness of this multi-model attribution analyses are needed to quantify the exact mechanism depends on the delicate balance between land- role of past and future AA on the weakening of storm tracks. atmosphere feedbacks and the strength (and thus its long-term There is still substantial uncertainty in what implications this weakening) of the storm tracks. weakening will have for summer weather conditions. Does it Recent studies indicate that the Atlantic Meridional Over- become more persistent and therefore more extreme? Theoreti- turning Circulation (AMOC, i.e., the large-scale north-south cally, a weakened westerly flow leads to a small increase in the transport in the Atlantic ocean) has seen an unprecedented stationary wavenumber (i.e., a 5% decrease in the flow would lead 103,104 slowdown in recent decades , something which is projected to a 2.5% increase in stationary wavenumber ) but not neces- for future warmer climates as well. This slowdown results in sarily to a more-stationary flow. Both observations and climate anomalously cold SSTs over the northern Atlantic which can models suggest that a weaker westerly flow is associated with a trigger a quasi-stationary Rossby wave response favoring blocking more wavy flow pattern, especially in summer, but this does not 60,61 high-pressure systems over western Europe . So, just like AA, a necessarily imply a cause-effect relationship (see Box 3). A slowdown of the AMOC leads to weakening westerlies in summer reduction in synoptic activity favors the buildup of hot-dry over the Atlantic sector, favoring persistent hot-dry extremes over conditions and this can interact with regional land-atmosphere Europe . Recent observational studies indeed indicate that processes that can feedback on the continental-scale circulation weather persistence in Europe and some other mid-latitude (as outlined above). However, the relative role of these processes, 23–25,102 regions has increased in boreal summer . and the strength of their interactions is largely unquantified. To 8 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE assess future high-end risks, these non-linear interactions need to the atmosphere. This can give direct insight into the underlying be disentangled and quantified. physical reasons behind model bias and thus provides concrete There is robust evidence that summer time quasi-stationary targets for model improvements. waves lead to persistent and therefore extreme weather condi- Finally, high-resolution paleo-climate records over the Holo- tions. However, substantial uncertainty remains in how such cene period can provide further insights into the circulation waves will change under global warming including the role of AA response to temperature gradient changes and put recent trends therein. While physical mechanisms exist that could amplify into a long-term perspective. Paleo proxies typically measure quasi-stationary short waves as an indirect response to AA, their some form of biological activity and are therefore best suited to representation in climate models is biased and their relative analyze the summer/growing season. The mid-Holocene provides importance compared to other drivers is poorly understood . a possible paleo analog with enhanced high-latitude warming Some upward trends in quasi-stationary wave activity have been and, interestingly, this period was also characterized by enhanced reported but confidence in these trends is generally low due to the drought conditions in the mid-latitudes . Likewise, tree-ring use of different diagnostics leading to conflicting results (Box 3). analyses suggest that enhanced jet stream waviness during sum- Moreover, the relative importance of enhanced high-latitude mers in recent decades has been unprecedented over the post- warming, strengthened monsoons, and a weakening AMOC with 1725 period . associated mid-latitude SST anomalies, in shaping the mid- latitude summer circulation needs to be understood. All these Summary drivers have a tendency to strengthen CGWTs but their inter- Future impacts from extreme weather are likely to be most pro- actions and regional effects are poorly known. nounced in summer, as most ecological activity and agricultural production takes place in this season . Though the uncertainties are large, changes in atmosphere dynamics have the potential to Ways forward cause rapid transitions at a regional scale leading to surprises for Given the societal risks and large uncertainties, we argue for society. In summer synergistic effects between thermodynamic coordinated research efforts to address the knowledge gaps and dynamic drivers of extreme weather could act in the same described above. To efficiently address this overarching theme, direction to cause very-extreme extremes . Recent summers have tighter collaboration between sub-disciplines within climate sci- seen such anomalous weather (Box 1) and these events are not ences is desirable, including scientists studying Arctic processes, well understood. This presents risks for society and in particular monsoons, storm track dynamics, (sub) seasonal forecasts, and for global food production, given that the major breadbasket extreme weather. regions are located in the mid-latitudes with many crop types Idealized atmosphere models are useful to study individual tel- vulnerable to heat extremes . econnections and their drivers. However, a central challenge is to The current literature provides robust evidence that AA quantify the interactions between Arctic teleconnections and other influences mid-latitude summer circulation substantially by teleconnections and regional processes, requiring state-of-the-art weakening the storm tracks. The uncertainties to do with other weather or climate models. Teleconnections are generally state- dynamical aspects and with how dynamical changes ultimately dependent (e.g., the Arctic’sinfluence might be pronounced only if affect regional weather conditions are admittedly large. thetropics areinaspecific state) and non-stationary. While often Nevertheless, we identified several possible feedback mechan- the mean circulation response to a certain perturbation (e.g., isms for how storm track weakening can lead to persistent and removing sea-ice) is analyzed, it will be important to quantify the therefore extreme weather in the mid-latitudes. Several studies response in probabilistic terms: The change in frequency of certain suggest that Northern Hemisphere summer weather is indeed 23,24,66,119 high-impact circulation regimes such as amplified and persistent already becoming more persistent . quasi-stationary waves. An increased frequency in such rare regimes In summary, this review shows that AA is likely to have sub- will have little influence on the mean circulation but can have large stantial impacts on mid-latitude summer circulation. The societal societal impacts. To do so will require large ensembles and well- impacts can be severe due to tail risks arising from radiatively coordinated experiments with a range of different models. To dis- forced mean summer warming combined with local and remote entangle the influence of different teleconnections, a so-called processes that favor more persistent summer weather. A coor- storyline approach can be insightful as well. This avoids quantifying dinated research agenda focusing on summer circulation, its probabilities associated with dynamical changes altogether and, drivers and extremes is needed to resolve the key knowledge gaps. instead, creates a set of physically plausible scenarios (i.e., storylines) 110,111 of future changes . For example, the change in European Received: 26 July 2017 Accepted: 20 June 2018 summer circulation could be described by combining a set of storylines that are based on the response to remote drivers from the Arctic, the tropics, AMOC, and regional soil-moisture changes. This way, the high-end risk (e.g., when Arctic and tropical teleconnec- tions combine with soil-moisture feedbacks to push European summer weather towards much more persistent hot-dry condi- References 1. Lehmann, J., Coumou, D. & Frieler, K. Increased record-breaking tions) can be identified and studied without assigning a specific precipitation events under global warming. Clim. Chang. 132, 501–515 (2015). probability to it . 2. Westra, S., Alexander, L. V. & Zwiers, F. W. Global increasing trends in To understand and overcome model biases, e.g., in the repre- annual maximum daily precipitation. J. Clim. 26, 3904–3918 (2013). sentation of summer Rossby waves and ocean-atmosphere 3. Fischer, E. M. & Knutti, R. Anthropogenic contribution to global occurrence feedbacks in the presence of sea-ice , novel machine learning of heavy-precipitation and high-temperature extremes. Nat. Clim. Chang. 5, approaches should be used to better integrate information from 560–564 (2015). 4. Coumou, D. & Rahmstorf, S. A decade of weather extremes. Nat. Clim. Chang. observations in climate models. In particular, causal discovery 2, 491–496 (2012). algorithms can identify causal pathways in the atmosphere and 5. Shepherd, T. G. Atmospheric circulation as a source of uncertainty in climate quantify their relative importance from observations alone . change projections. Nat. Geosci. 7, 703–708 (2014). They can thus be used to do process-based model validations and 6. Cohen, J. et al. Recent Arctic amplification and extreme mid-latitude weather. quantify how well models represent certain causal pathways in Nat. Geosci. 7, 627–637 (2014). NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 9 REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 7. Screen, J. A. Climate science: far-flung effects of Arctic warming. Nat. Geosci. 38. Schubert, S., Wang, H. & Suarez, M. Warm season subseasonal variability and 10, 253–254 (2017). climate extremes in the Northern Hemisphere: The role of stationary Rossby 8. Vihma, T. Weather extremes linked to interaction of the Arctic and waves. J. Clim. 24, 4773–4792 (2011). Midlatitudes. Clim. Extremes 226,39–50 (2017). 39. Branstator, G. Circumglobal teleconnections, the jet stream waveguide, and 9. Kretschmer, M., Coumou, D., Tziperman, E. & Cohen, J. More-persistent the North Atlantic oscillation. J. Clim. 15, 1893–1910 (2002). weak stratospheric polar vortex states linked to cold extremes. Bull. Am. 40. Trenberth, K. E. et al. Progress during TOGA in understanding and modeling Meteor. Soc. 99,49–60 (2018). global teleconnections associated with tropical sea surface temperatures. 10. Budikova, D. Role of Arctic sea ice in global atmospheric circulation: a review. J. Geophys. Res. 103, 14291–14324 (1998). Glob. Planet. Chang. 68, 149–163 (2009). 41. Yu, J.-Y., Wang, X., Yang, S., Paek, H. & Chen, M. The changing El Nino- 11. Overland, J. E. et al. Nonlinear response of mid-latitude weather to the Southern oscillation and associated climate extremes. Clim. Extremes 226, changing Arctic. Nat. Clim. Chang. 6, 992–999 (2016). 3–38 (2017). 12. Walsh, J. E. Intensified warming of the Arctic: causes and impacts on middle 42. Eden, J. M., Oldenborgh, G. J., Van, Hawkins, E. & Suckling, E. B. A global latitudes. Glob. Planet. Chang. 117,52–63 (2014). empirical system for probabilistic seasonal climate. Geosci. Model Dev. 8, 13. Francis, J. A., Vavrus, S. J. & Cohen, J. Amplified Arctic warming and mid- 3947–3973 (2015). latitude weather: new perspectives on emerging connections. WIREs Clim. 43. Weisheimer, A. & Palmer, T. N. On the reliability of seasonal climate Chang. 8, e474 (2017). forecasts. J. R. Soc. Interface 11, 20131162 (2014). 14. Vihma, T. Effects of Arctic sea ice decline on weather and climate: a review. 44. Ding, Q. & Wang, B. Circumglobal teleconnection in the Northern Surv. Geophys. 35, 1175–1214 (2014). Hemisphere Summer*. J. Clim. 18, 3483–3505 (2005). 15. Shepherd, T. G. Effects of a warming Arctic. Science 353, 989–990 (2016). 45. Ding, Q. & Wang, B. Intraseasonal teleconnection between the summer 16. Cohen, J. et al. Arctic change and possible influence on mid-latitude climate Eurasian wave train and the Indian Monsoon. J. Clim. 20, 3751–3767 and weather. US CLIVAR White Paper. https://doi.org/10.5065/ (2007). D6TH8KGW (2018). 46. Shaw, T. A. & Voigt, A. Tug of war on summertime circulation between 17. Horton, R. M., Mankin, J. S., Lesk, C., Coffel, E. & Raymond, C. A review of radiative forcing and sea surface warming. Nat. Geosci. 8, 560–566 (2015). recent advances in research on extreme heat events. Curr. Clim. Chang. Rep. 2, 47. Hall, R. J., Jones, J. M., Hanna, E. & Scaife, A. A. Drivers and potential 242–259 (2016). predictability of summer time North Atlantic polar front jet variability. Clim. 18. Schär, C. et al. The role of increasing temperature variability in European Dyn. 48, 3869–3887 (2017). summer heatwaves. Nature 427, 332–336 (2004). 48. Derksen, C. & Brown, R. Spring snow cover extent reductions in the 2008- 19. Petoukhov, V., Rahmstorf, S., Petri, S. & Schellnhuber, H. J. Quasiresonant 2012 period exceeding climate model projections. Geophys. Res. Lett. 39, amplification of planetary waves and recent Northern Hemisphere weather L19504 (2012). extremes. Proc. Natl Acad. Sci. USA 110, 5336–5341 (2013). 49. Kunkel, K. E. et al. Trends and extremes in Northern Hemisphere snow 20. Russo, S. et al. Magnitude of extreme heat waves in present climate and their characteristics. Curr. Clim. Chang. Rep. 2,65–73 (2016). projection in a warming world. J. Geophys. Res. Atmos. 119,500–12,512 (2014). 50. Wu, R., Zhao, P. & Liu, G. Change in the contribution of spring snow cover 21. Christidis, N., Jones, G. S. & Stott, P. A. Dramatically increasing chance of and remote oceans to summer air temperature anomaly over Northeast China extremely hot summers since the 2003 European heatwave. Nat. Clim. Chang. around 1990. J. Geophys. Res. Atmos. 119, 663–676 (2014). 5,46–50 (2014). 51. Halder, S. & Dirmeyer, P. A. Relation of Eurasian snow cover and Indian 22. Weaver, S. J., Kumar, A. & Chen, M. in Climate Extremes: Patterns and summer monsoon rainfall: Importance of the delayed hydrological effect. Mechanisms, Geophysical Monograph 226 1st edn (eds Simon Wang, S.-Y. J. Clim. 30, 1273–1289 (2017). et al.) 105–114 (Wiley, New York, 2017). 52. Alexander, M., Tomas, R., Deser, C. & Lawrence, D. The atmospheric 23. Pfleiderer, P. & Coumou, D. Quantification of temperature persistence over response to projected terrestrial snow changes in the late twenty-first century. the Northern Hemisphere land-area. Clim. Dyn. 51, 627–637 (2018). J. Clim. 23, 6430–6437 (2010). 24. Hoffmann, P. Enhanced seasonal predictability of the summer mean 53. Matsumura, S. & Yamazaki, K. Eurasian Subarctic summer climate in temperature in Central Europe caused by new dominant weather patterns. response to anomalous snow cover. J. Clim. 25, 1305–1317 (2012). Clim. Dyn. 50, 2799–2812 (2018). 54. Hoskins, B. J. & Karoly, D. J. The steady linear response of a spherical 25. Horton, D. E. et al. Contribution of changes in atmospheric circulation atmosphere to thermal and orographic forcing. J. Atmos. Sci. 38, 1179–1196 patterns to extreme temperature trends. Nature 522, 465–469 (2015). (1981). 26. Coumou, D., Lehmann, J. & Beckmann, J. The weakening summer circulation 55. Hoskins, B. J. & Ambrizzi, T. Rossby wave propagation on a realistic in the Northern Hemisphere mid-latitudes. Science 348, 324–327 (2015). longitudinally varying flow. J. Atmos. Sci. 50, 1661–1671 (1983). 27. Barnes, E. A. & Polvani, L. Response of the midlatitude jets, and of their 56. Teng, H., Branstator, G., Wang, H., Meehl, G. A. & Washington, W. M. variability, to increased greenhouse gases in the CMIP5 models. J. Clim. 26, Probability of US heat waves affected by a subseasonal planetary wave pattern. 7117–7135 (2013). Nat. Geosci. 6, 1056–1061 (2013). 28. Barnes, E. A. & Polvani, L. M. CMIP5 projections of Arctic amplification, of 57. Stadtherr, L., Coumou, D., Petoukhov, V., Petri, S. & Rahmstorf, S. Record the North American/North Atlantic circulation, and of their relationship. Balkan floods of 2014 linked to planetary wave resonance. Sci. Adv. 2, J. Clim. 28, 5254–5271 (2015). e1501428 (2016). 29. Harvey, B. J., Shaffrey, L. C. & Woollings, T. J. Equator-to-pole temperature 58. Wang, S. Y., Hipps, L., Gillies, R. R. & Yoon, J. H. Probable causes of the differences and the extra-tropical storm track responses of the CMIP5 climate abnormal ridge accompanying the 2013–2014 California drought: ENSO models. Clim. Dyn. 43, 1171–1182 (2014). precursor and anthropogenic warming footprint. Geophys. Res. Lett. 41, 30. Harvey, B. J., Shaffrey, L. C. & Woollings, T. J. Deconstructing the climate 3220–3226 (2014). change response of the Northern Hemisphere wintertime storm tracks. Clim. 59. McKinnon, K. A., Rhines, A., Tingley, M. P. & Huybers, P. Long-lead Dyn. 45, 2847–2860 (2015). prediction of eastern United States hot days from Pacific sea surface 31. Zappa, G., Pithan, F. & Shepherd, T. G. Multi-model evidence for an temperatures. Nat. Geosci. 9, 389–394 (2016). atmospheric circulation response to Arctic sea ice loss in the CMIP5 future 60. Vavrus, S. J. et al. Changes in North American atmospheric circulation and projections. Geophys. Res. Lett. 45, 1011–1019 (2018). extreme weather: influence of Arctic amplification and Northern Hemisphere 32. Perlwitz, J., Hoerling, M. & Dole, R. Arctic tropospheric warming: causes and snow cover. J. Clim. 30, 4317–4333 (2017). linkages to lower latitudes. J. Clim. 28, 2154–2167 (2015). 61. Cattiaux, J., Peings, Y., Saint-Martin, D., Trou-Kechout, N. & Vavrus, S. J. 33. Petoukhov, V. & Semenov, V. A. A link between reduced Barents-Kara sea ice Sinuosity of midlatitude atmospheric flow in a warming world. Geophys. Res. and cold winter extremes over northern continents. J. Geophys. Res. 115, Lett. 43, 8259–8268 (2016). D21111 (2010). 62. Di Capua, G. & Coumou, D. Changes in meandering of the Northern 34. Kim, B.-M. et al. Weakening of the stratospheric polar vortex by Arctic sea-ice Hemisphere circulation. Environ. Res. Lett. 11,1–9 (2016). loss. Nat. Commun. 5, 4646 (2014). 63. Hoskins, B. & Woollings, T. Persistent extratropical regimes and climate 35. Jaiser, R., Dethloff, K., Handorf, D., Rinke, A. & Cohen, J. Impact of sea ice extremes. Curr. Clim. Chang. Rep. 1, 115–124 (2015). cover changes on the Northern Hemisphere atmospheric winter circulation. 64. Petoukhov, V. et al. Role of quasiresonant planetary wave dynamics in recent Tellus A 64,1–11 (2012). boreal spring-to-autumn extreme events. Proc. Natl Acad. Sci. USA 113, 36. Kretschmer, M., Coumou, D., Runge, J. & Donges, J. Using causal effect 6862–6867 (2016). networks to analyze different Arctic drivers of mid-latitude winter circulation. 65. Coumou, D., Petoukhov, V., Rahmstorf, S., Petri, S. & Schellnhuber, H. J. J. Clim. 29, 4069–4081 (2016). Quasi-resonant circulation regimes and hemispheric synchronization of 37. Kidston, J. et al. Stratospheric influence on tropospheric jet streams, storm extreme weather in boreal summer. Proc. Natl Acad. Sci. USA 111, tracks and surface weather. Nat. Geosci. 8, 433–440 (2015). 12331–12336 (2014). 10 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 REVIEW ARTICLE 66. Kornhuber, K., Petoukhov, V., Petri, S., Rahmstorf, S. & Coumou, D. Evidence 97. Vavrus, S. The role of terrestrial snow cover in the climate system. Clim. Dyn. for wave resonance as a key mechanism for generating high-amplitude quasi- 29,73–88 (2007). stationary waves in boreal summer. Clim. Dyn. 49, 1961–1979 (2017). 98. Matsumura, S., Zhang, X. & Yamazaki, K. Summer Arctic atmospheric 67. Reed, R. J. & Kunkel, B. The Arctic circulation in summer. J. Meteorol. 17, circulation response to spring Eurasian snow cover and its possible linkage to 489–506 (1960). accelerated sea ice decrease. J. Clim. 25, 1305–1317 (2014). 68. Manola, I., Selten, F., De Vries, H. & Hazeleger, W. ‘Waveguidability’ of 99. Fischer, E. M., Seneviratne, S. I., Vidale, P. L., Lüthi, D. & Schär, C. Soil idealized jets. J. Geophys. Res. Atmos. 118, 10432–10440 (2013). moisture-atmosphere interactions during the 2003 European summer heat 69. Kornhuber, K. et al. Summertime planetary wave-resonance in the Northern wave. J. Clim. 20, 5081–5099 (2007). and Southern Hemisphere. J. Clim. 30, 6133–6150 (2017). 100. Haarsma, R. J. et al. Drier Mediterranean soils due to greenhouse warming 70. Tachibana, Y., Nakamura, N., Komiya, H. & Takahashi, M. Abrupt evolution bring easterly winds over summertime central Europe. Geophys. Res. Lett. 36, of the summer Northern Hemisphere annular mode and its association with 1–7 (2009). blocking. J. Geophys. Res. 115, D12125 (2010). 101. Bladé, I., Liebmann, B., Fortuny, D. & van Oldenborgh, G. J. Observed and 71. Chang, E. K. M., Ma, C., Zheng, C. & Yau, A. M. W. Observed and projected simulated impacts of the summer NAO in Europe: implications for projected decrease in Northern Hemisphere extratropical cyclone activity in summer and drying in the Mediterranean region. Clim. Dyn. 39,709–727 (2012). its impacts on maximum temperature. Geophys. Res. Lett. 43,2200–2208 (2016). 102. Alvarez-Castro, M. C., Faranda, D. & Yiou, P. Atmospheric dynamics leading 72. Petrie, R. E., Shaffrey, L. C. & Sutton, R. T. Atmospheric response in summer to West European summer hot temperatures since 1851. linked to recent Arctic sea ice loss. Q. J. R. Meteorol. Soc. 141, 2070–2076 (2015). Complexity 2018, 2494509 (2018). 73. Lee, S. A theory of polar amplification from a general circulation perspective. 103. Rahmstorf, S. et al. Exceptional twentieth-century slowdown in Atlantic J. Atmos. Sci., 50,31–43 (2014). Ocean overturning circulation. Nat. Clim. Chang. 5, 475–480 (2015). 74. Coumou, D., Kornhuber, K., Lehmann, J. & Petoukhov, V. Weakened flow 104. Caesar, L., Rahmstorf, S., Robinson, A. G. F. & Saba, V. Observed fingerprint of persistent circulation, and prolonged weather extremes in boreal summer. a weakening Atlantic Ocean overturning circulation. Nature 556, 191–196 Clim. Extremes 226,61–73 (2017). (2018). 75. Knudsen, E., Orsolini, Y., Furevik, T. & Hodges, K. Observed anomalous 105. Duchez, A., Frajka-Williams, E., Josey, S. A., Evans, D. G. & Grist, J. P. Drivers atmospheric circulation in summers of unusual Arctic Sea ice reduction. of exceptionally cold North Atlantic Ocean temperatures and their link to the J. Geophys. Res. 16, 2014 (2014). 2015 European heat wave. Environ. Res. Lett. 11, 074004 (2016). 76. Mann, M. E. et al. Influence of anthropogenic climate change on planetary 106. Haarsma, R. J., Selten, F. M. & Drijfhout, S. S. Decelerating Atlantic wave resonance and extreme weather events. Sci. Rep. 7, 45242 (2017). meridional overturning circulation main cause of future west European 77. Blackport, R. & Kushner, P. J. The transient and equilibrium climate response summer atmospheric circulation changes. Environ. Res. Lett. 10, 94007 (2015). to rapid summertime sea ice loss in CCSM4. J. Clim. 29, 401–417 (2016). 107. Kripalani, R. H., Kulkarni, A. & Sabade, S. S. El Nino southern oscillation, 78. Screen, J. A. Influence of Arctic sea ice on European summer precipitation. Eurasian snow cover and the indian monsoon rainfall. Indian Natl Sci. Acad. Environ. Res. Lett. 8, 044015 (2013). 67, 361–368 (2001). 79. Lehmann, J., Coumou, D., Frieler, K., Eliseev, A. V. & Levermann, A. Future 108. Vellore, R. K. et al. Monsoon—extratropical circulation interactions in changes in extratropical storm tracks and baroclinicity under climate change. Himalayan extreme rainfall. Clim. Dyn. 46, 3517–3546 (2016). Environ. Res. Lett. 9, 084002 (2014). 109. Tang, Q., Zhang, X. & Francis, J. A. Extreme summer weather in northern 80. Brewer, M. C. & Mass, C. F. Projected changes in Western U. S. large-scale mid-latitudes linked to a vanishing cryosphere. Nat. Clim. Chang. 4,45–50 summer synoptic circulations and variability in CMIP5 models. J. Clim. 29, (2013). 5965–5978 (2016). 110. Zappa, G. & Shepherd, T. G. Storylines of atmospheric circulation change for 81. Deser, C., Knutti, R., Solomon, S. & Phillips, A. S. Communication of the role European regional climate impact assessment. J. Clim. 30, 6561–6577 (2017). of natural variability in future North American climate. Nat. Clim. Chang. 2, 111. Hazeleger, W. et al. Tales of future weather. Nat. Clim. Chang. 5, 107–113 775–779 (2012). (2015). 82. Lu, J., Chen, G. & Frierson, D. The position of the midlatitude storm track and 112. Overland, J. E., Francis, J. A., Hanna, E. & Wang, M. The recent shift in early Eddy-driven westerlies in aquaplanet AGCMs. J. Atmos. Sci. 67, 3984–4000 summer Arctic atmospheric circulation. Geophys. Res. Lett. 39, L19804 (2010). (2012). 83. Grise, K. & Polvani, L. M. The response of midlatitude jets to increased CO : 113. Pithan, F. et al. Select strengths and biases of models in representing the Arctic distinguishing the roles of sea surface temperature and direct radiative forcing. winter boundary layer over sea ice: the Larcform 1?single column model Geophys. Res. Lett. 41, 6863–6871 (2014). intercomparison. J. Adv. Model. Earth Syst. 8, 764–785 (2016). 84. Butler, A. H., Thompson, D. W. J. & Heikes, R. The steady-state atmospheric 114. Runge, J. et al. Identifying causal gateways and mediators in complex spatio- circulation response to climate change–like thermal forcings in a simple temporal systems. Nat. Commun. 6, 8502 (2015). general circulation model. J. Clim. 23, 3474–3496 (2010). 115. Routson, C. et al. Changing temperature gradients linked to Holocene 85. Newson, R. L. Response of a general circulation model of the atmosphere to moisture trends in the Northern Hemisphere. In American Geophysical Union, removal of the Arctic ice-cap. Nature 241,39–40 (1973). Fall General Assembly 2016 (AGU, 2016). 86. Vallis, G. K., Zurita-Gotor, P., Cairns, C. & Kidston, J. Response of the large- 116. Trouet, V., Babst, F. & Meko, M. Recent enhanced high-summer North scale structure of the atmosphere to global warming. Q. J. R. Meteorol. Soc. Atlantic Jet variability emerges from three-century context. Nat. Commun. 9, 141, 1479–1501 (2015). 180 (2018). 87. Palmer, T. N. Nonlinear dynamics and climate change: Rossby’s legacy. Bull. 117. Hansen, J., Sato, M. & Ruedy, R. Perception of climate change. Proc. Natl Am. Meteorol. Soc. 79, 1411–1423 (1998). Acad. Sci. USA 109, E2415–E2423 (2012). 88. Molnos, S. et al. A network-based detection scheme of the jet stream core. 118. Lobell, D. B., Schlenker, W. & Costa-Roberts, J. Climate trends and global crop Earth Syst. Dyn. 8,75–89 (2017). production since 1980. Science 333, 616–620 (2011). 89. Archer, C. L. & Caldeira, K. Historical trends in the jet streams. Geophys. Res. 119. Kossin, J. P. A global slowdown of tropical-cyclone translation speed. Lett. 35,1–6 (2008). Nature 558, 104–108 (2018). 90. Fu, Q. & Lin, P. Poleward shift of subtropical jets inferred from satellite- 120. Fannin, B. Updated 2011 Texas agricultural drought losses total $7.62 observed lower-stratospheric temperatures. J. Clim. 24, 5597–5603 (2011). billion. AgriLifeTODAY March 21. http://today.agrilife.org/2012/03/21/ 91. Bender, F. A.-M., Ramanathan, V. & Tselioudis, G. Changes in extratropical updated-2011-texas-agricultural-drought-losses-total-7-62-billion/ (2012). storm track cloudiness 1983-2008: observational support for a poleward shift. 121. Rippey, B. R. The U.S. drought of 2012. Wea. Clim. Extremes 10,57–64 Clim. Dyn. 38, 2037–2053 (2011). (2015). 92. Zappa, G., Lucarini, V. & Navarra, A. Baroclinic stationary waves in 122. Zscheischler, J. & Seneviratne, S. I. Dependence of drivers affects risks aquaplanet models. J. Atmos. Sci. 68, 1023–1040 (2011). associated with compound events. Sci. Adv. 3, e1700263 (2017). 93. Wang, S. Y. S. et al. An intensified seasonal transition in the Central U.S. that 123. Leonard, M. et al. A compound event framework for understanding extreme enhances summer drought. J. Geophys. Res. 120, 8804–8816 (2015). impacts. WIREs Clim. Change 5, 113–128 (2014). 94. Lee, M. H., Lee, S., Song, H.-Y. & Ho, C.-H. The recent increase in the 124. Koster, R. D. et al. GLACE Team, Regions of strong coupling between soil occurrence of a boreal summer teleconnection and its relationship with moisture and precipitation. Science 305, 1138–1140 (2004). temperature extremes. J. Clim. 30, 7493–7504 (2017). 125. Koster, R. D., Chang, Y., Wang, H., & Schubert, S. D. Impacts of local soil 95. Wang, S. Y., Davies, R. E. & Gillies, R. R. Identification of extreme moisture anomalies on the atmospheric circulation and on remote surface precipitation threat across midlatitude regions based on short-wave meteorological fields during boreal summer: A comprehensive analysis over circulations. J. Geophys. Res. Atmos. 118, 11059–11074 (2013). North America. J. Clim. 29, 7345–7364 (2016). 96. Masato, G., Hoskins, B. & Woollings, T. Winter and summer Northern 126. Deser, C., Terray, L. & Phillips, A. S. Forced and internal components of hemisphere blocking in CMIP5 models. J. Clim. 26, 7044–7059 winter air temperature trends over North America during the past 50 years: (2013). Mechanisms and implications. J. Clim. 29, 2237–2258 (2016). NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications 11 REVIEW ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-018-05256-8 127. Lehner, F., Deser, C. & Terray, L. Toward a new estimate of ‘‘time of 144. Huang, C. S. & Nakamura, N. Local finite-amplitude wave activity as a emergence’’ of anthropogenic warming: insights from dynamical adjustment diagnostic of anomalous weather events. J. Atmos. Sci. 73, 211–229 (2016). and a large initial-condition model ensemble. J. Clim. 30, 7739–7756 (2017). Acknowledgements 128. Merrifield, A., Lehner, F., Xie, S. P. & Deser, C. Removing circulation effects to This work was supported by the German Federal Ministry of Education and Research assess central U.S. land-atmosphere interactions in the CESM Large (BMBF), grant 01LP1611A (G.D.C. and D.C.) and the Netherlands Organisation for Ensemble. Geophys. Res. Lett. 44, 9938–9946 (2017). Scientific Research (NWO), grant 016.Vidi.171011 (D.C.). S.W. is supported by the the 129. Bandyopadhyay, S., Kanji, S. & Wang, L. The impact of rainfall and U.S. Department of Energy Office of Science, Biological and Environmental Research temperature variation on diarrheal prevalence in Sub-Saharan Africa. Appl. program (BER) grant DESC0016605 and the Utah Agricultural Experiment Station, Utah Geogr. 33,63–72 (2012). State University, with partial support from the Central Weather Bureau of Taiwan. 130. Allen, C. D. et al. A global overview of drought and heat-induced tree Further support was provided by the U.S. National Science Foundation, grants PLR- mortality reveals emerging climate change risks for forests. For. Ecol. 1304398 and AGS-1640452 (S.V.). L.W. was supported by NASA grant 80NSSC17K0267, Management 259, 660–684 (2010). NSF grant AGS-1552385, and a grant from the Harvard Global Institute. 131. Flannigan, M. D., Krawchuk, M. A., de Groot, W. J., Wotton, B. M. & Gowman, L. M. Implications of changing climate for global wildland fire. Int. J. Wildland Fire 18, 483–507 (2009). 132. Anyamba, A. et al. Recent weather extremes and impacts on agricultural Author contributions production and vector-borne disease outbreak patterns. PLOS ONE 9, e92538 D.C. initiated this review paper and developed the concept, and all authors (D.C., G.D.C., (2014). S.V., L.W. and S.W.) contributed equally to the writing of the manuscript. 133. Donat, M. G. et al. Global land-based datasets for monitoring climatic extremes. Bull. Am. Meteorol. Soc. 94, 997–1006 (2013). Additional information 134. Kalnay, E. et al. The NCEP/NCAR 40-year reanalysis project. Bull. Am. Competing interests: The authors declare no competing interests. Meteorol. Soc. 77, 437–471 (1996). 135. Hurrell, J. W. Decadal trends in the North Atlantic Oscillation regional Reprints and permission information is available online at http://npg.nature.com/ temperatures and precipitation. Science 269, 676–679 (1995). reprintsandpermissions/ 136. Thompson, D. W. J. & Wallace, J. M. Regional climate impacts of the Northern Hemisphere annular mode. Science 293,85–89 (2001). Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in 137. Ogi, M., Yamazaki, K. & Tachibana, Y. The summertime annular mode in the published maps and institutional affiliations. Northern Hemisphere and its linkage to the winter mode. J. Geophys. Res. D. Atmos. 109,1–15 (2004). 138. Folland, C. K. et al. The summer North Atlantic oscillation: past, present, and future. J. Clim. 22, 1082–1103 (2009). Open Access This article is licensed under a Creative Commons 139. Pfahl, S., Schwierz, C., Croci-Maspoli, M., Grams, C. M. & Wernli, H. Attribution 4.0 International License, which permits use, sharing, Importance of latent heat release in ascending air streams for atmospheric adaptation, distribution and reproduction in any medium or format, as long as you give blocking. Nat. Geosci. 8, 610–614 (2015). appropriate credit to the original author(s) and the source, provide a link to the Creative 140. Rohli, R. V., Wrona, K. M. & Mchugh, M. J. January northern hemisphere Commons license, and indicate if changes were made. The images or other third party circumpolar vortex variability and its relationship with hemispheric temperature material in this article are included in the article’s Creative Commons license, unless and regional teleconnections. Int. J. Climatol. 25,1421–1436 (2005). indicated otherwise in a credit line to the material. If material is not included in the 141. Barnes, E. A. Revisiting the evidence linking Arctic amplification to extreme article’s Creative Commons license and your intended use is not permitted by statutory weather in midlatitudes. Geophys. Res. Lett. 40, 4734–4739 (2013). regulation or exceeds the permitted use, you will need to obtain permission directly from 142. Nakamura, N. Two-dimensional mixing, edge formation, and permeability the copyright holder. To view a copy of this license, visit http://creativecommons.org/ diagnosed in an area coordinate. J. Atmos. Sci. 53, 1524–1537 licenses/by/4.0/. (1996). 143. Nakamura, N. & Zhu, D. Finite-amplitude wave activity and diffusive flux of potential vorticity in Eddy–Mean flow interaction. J. Atmos. Sci. 67, © The Author(s) 2018 2701–2716 (2010). 12 NATURE COMMUNICATIONS | (2018) 9:2959 | DOI: 10.1038/s41467-018-05256-8 | www.nature.com/naturecommunications

Journal

Nature CommunicationsSpringer Journals

Published: Aug 20, 2018

There are no references for this article.