Fig. 1: The dependence of SST and surface heat-flux on AMV phase. | npj Climate and Atmospheric Science

Fig. 1: The dependence of SST and surface heat-flux on AMV phase.

From: Challenges with interpreting the impact of Atlantic Multidecadal Variability using SST-restoring experiments

Fig. 1

a–i SST difference between SST+ and SST- 10-year periods for the AMV, AMV-Trop and AMV-ExTrop regions (indicated by the black boxes). These are shown for the SST-restoring experiments for the IPSL and UM models; for the free-running coupled PiControl simulations the SST+ and SST- decades are decades in which the SST in the region is above/below one standard deviation from the mean. j–r Surface heat-flux, Qnet (defined as positive upwards), difference between SST+ and SST- 10-year periods for the AMV, AMV-Trop and AMV-ExTrop regions. Hatching indicates where the differences are not significant at the 95% level, based on a Monte Carlo resampling performed 10000 times (see Methods). s Decadal SST anomalies (HadISST, 1870-2014) and (t) Qnet anomalies (20CR, 1870–2014) regressed onto a normalised decadal SST index averaged over the AMV region. Hatching in (s, t) indicates where the regression coefficients are not significant at the 95% level, based on a Monte Carlo phase randomisation test (see Methods). The green boxes indicate the midlatitude and subtropical North Atlantic regions used in Fig. 2.

Back to article page