Content area
Full text
1. Introduction
Tropical Pacific sea surface temperatures (SSTs) influence midlatitude atmospheric variability through hemispheric-spanning teleconnection patterns [Exner 1914; Horel and Wallace 1981; Hurrell 1996; Trenberth et al. 1998; Trenberth and Caron 2000; Alexander et al. 2002; Ciasto and Thompson 2008; Frauen et al. 2014; Deser et al. 2017; and see extended reviews in Hoerling and Kumar (2002) and Stan et al. (2017)]. The dominant mode of tropical Pacific SST variability, El Niño–Southern Oscillation (ENSO), affects weather and climate over much of the globe even in high-latitude regions with potential impacts on surface temperature trends over Greenland (Ding et al. 2014), Arctic sea ice variability (Wettstein and Deser 2014; Ding et al. 2017), and heat transport into polar regions (Park et al. 2015). In the North Atlantic–European sector, however, determining the robust signals of ENSO has been a challenge (Brönnimann et al. 2007). During El Niño winters, there is a tendency for negative North Atlantic Oscillation (NAO)-like conditions (i.e., equatorward-shifted North Atlantic jet, lower than normal temperatures over northern Europe; increased precipitation over southern Europe), and vice versa during La Niña (van Loon and Madden 1981; May and Bengtsson 1998; Moron and Plaut 2003; Brönnimann et al. 2007; García-Serrano et al. 2011; Li and Lau 2012a; Rodríguez-Fonseca et al. 2016) although there is some debate over whether ENSO truly triggers the NAO (García-Serrano et al. 2011).
A classical view on how these teleconnection signals are transmitted is via stationary Rossby wave propagation that alters the time-mean extratropical flow (Hoskins and Karoly 1981; Hoskins and Ambrizzi 1993; Honda et al. 2001; Held et al. 2002; Moron and Gouirand 2003). The Rossby waves follow great-circle pathways from the tropics into the midlatitudes (Hoskins and Karoly 1981; Hoskins and Ambrizzi 1993), along ray paths that depend on the midlatitude flow (Branstator 1985). In this way, tropical Pacific SST variability is assumed to drive changes in midlatitude storm tracks (e.g., May and Bengtsson 1998; Moron and Plaut 2003; Eichler and Higgins 2006; Ciasto et al. 2016; Branstator 2014) and consequently regional climate, for example, seasonally averaged precipitation and surface temperatures over North America (e.g., Ropelewski and Halpert 1987; Halpert and Ropelewski 1992; Gershunov and Barnett 1998; Seager et al. 2010; Smith and Sardeshmukh 2000; Yu et al. 2012).
This classical...