Fig. 5: Relationship between SIPNet skill and SIC variability under different ENSO phases.
From: ENSO’s impact on linear and nonlinear predictability of Antarctic sea ice
a Scatterplots of the impact of ENSO events on SIPNet skill as a function of the impact of ENSO events on SIC weekly variability within 1° longitude bins. The impact of El Niño (La Niña) on SIPNet skill is computed as the difference between ACC over 4–6-week leads under El Niño (La Niña) and neutral conditions, divided by the latter. The red, green, and blue dots represent the prediction skill corresponding to lead times of 4, 5, and 6 weeks, respectively. Similarly, The impact of El Niño (La Niña) on SIC weekly variability is computed as the difference between SIC weekly standard deviation under El Niño (La Niña) and neutral conditions, divided by the latter. b is similar to a but shows the regional averages. The solid line represents the linear fit of the data. “r” denotes the correlation coefficient, and “P” denotes the p value. “Weddell-El Niño” denotes the impact of El Niño events on sea ice variability and predictability in the Weddell Sea.
