There is medium confidence that the westerly winds in the Southern Ocean have increased since the early 1980s (Section 3.4.4), associated with a positive trend in the SAM (Marshall, 2003); also see Sections 3.4.4 and 3.6.3). Although a few observational studies have found evidence for correlation between SAM and ACC transport on subseasonal to interannual scales (e.g., Hughes et al., 2003; Meredith et al., 2004), there is no significant observational evidence of an increase in ACC transport associated with the multi-decadal trend in wind forcing over the Southern Ocean. Repeat hydrographic sections spread unevenly over 35 years in Drake Passage (e.g., Cunningham et al., 2003; Koshlyakov et al., 2007, 2011; Gladyshev et al., 2008), south of Africa (Swart et al., 2008) and south of Australia (Rintoul et al., 2002) reveal moderate variability but no significant trends in these sparse and discontinuous records. A comparison of recent Argo data and a long-term climatology showed that the slope of density surfaces (hence baroclinic transport) associated with the ACC had not changed in recent decades (Böning et al., 2008). Eddy-resolving models suggest the ACC transport is relatively insensitive to trends in wind forcing, consistent with the ACC being in an “eddy-saturated” state where increases in wind forcing are compensated by changes in the eddy field (Hallberg and Gnanadesikan, 2006; Farneti et al., 2010; Spence et al., 2010). While there is limited evidence for (or against) multi-decadal changes in transport of the ACC, observations of changes in temperature, salinity and SSH indicate the current system has shifted poleward (medium confidence) (Böning et al., 2008; Gille, 2008; Morrow et al., 2008; Sokolov and Rintoul, 2009; Kazmin, 2012).