Fig. 4: Causal links between the ISMR and AMO in the presence of other potential drivers using the Granger causal inference algorithm. | npj Climate and Atmospheric Science

Fig. 4: Causal links between the ISMR and AMO in the presence of other potential drivers using the Granger causal inference algorithm.

From: Predictability of South-Asian monsoon rainfall beyond the legacy of Tropical Ocean Global Atmosphere program (TOGA)

Fig. 4

Similar to Fig. 3a but the causal inference links are obtained from monthly anomalies of the same indices using a linear Granger causality and b a nonlinear Granger causality framework at 95% alpha level. Monthly anomalies of SST are computed from COBE SST2, Mean sea level pressure computed from NCEP 20Cv3 and ISMR from Parthasarathy data, spanning from 1871 to 2017 for MJJASO season. As in PCMCI+, both the linear and nonlinear causality are shown as arrows, with strength of association represented by arrow color (+ve red and –ve blue), while node color represents the node auto correlations. The curved lines represent a time-lagged causal relation, and the numbers denote the lag (in months).

Back to article page