The trivariate seasonal analysis of couplings between El Niño, North Atlantic Oscillation, and Indian monsoon

Abstract

Interdependencies between the El Niño-Southern Oscillation (ENSO), North Atlantic Oscillation (NAO), and Indian monsoon (IM) phenomena are investigated from data over the period of 1871-2013. Along with the bivariate analysis of directional couplings between ENSO, NAO, and IM, their trivariate analysis was carried out. To detect the seasonal features of directional couplings using Wiener-Granger causality, various temporal resolutions of the data ranging from a month to a half-year are used. Taking into account the seasonality of the processes, the influence of ENSO on NAO is detected which has different signs in winter and summer. The influence of NAO on ENSO is revealed only in the trivariate analysis. The strongest couplings are observed between ENSO and IM. All the detected couplings can be divided into two groups: the "fast" (with the characteristic time from a month to several months) and "slow" (with the characteristic time of a half-year or longer) ones. The fast couplings include bidirectional couplings between IM and ENSO in summer and autumn when the summer pattern of NAO influences the patterns of ENSO and IM in the next season. The slow couplings include the effects of ENSO on NAO and on the winter pattern of IM as well as the influence of IM on the summer pattern NAO.