Identifying quasi-periodic variability using multivariate empirical mode decomposition: a case of the tropical Pacific

Abstract

Abstract. Tropical Pacific is home to climate variability on different timescales, including El Niño Southern Oscillation (ENSO) – one of the most prominent quasi-periodic modes of variability in the Earth’s climate system. It is a coupled atmosphere-ocean mode of variability with a 2-8-year-timescale and oscillates between a warm (El Niño) and a cold (La Niña) phase. However, the dynamics of ENSO is complex, involving a variety of spatial and temporal scales as well as their interactions, which are not necessarily well understood. We use a recently developed nonlinear and nonstationary multivariate timeseries analysis tool – multivariate empirical mode decomposition (MEMD) – to revisit quasi-periodic variability within ENSO. MEMD is a powerful tool for objectively identifying (intrinsic) timescales of variability within a given system. We apply it to reanalysis and observational data as well as to climate model output (NorCPM1). Observational/reanalysis data reveal a quasi-periodic variability in the tropical Pacific on timescales ~2–4.5 years. This variability can then be related to ENSO’s recharge-discharge and simplified West-Pacific oscillator conceptual models. The latter occurs only on this timescale and is not necessarily well represented in NorCPM1. Additionally, the ~2–4.5-year variability in ENSO can be 'predicted' up to ~20 months ahead, while predicting the full ENSO amplitude remains challenging. MEMD can therefore be used for assessing climate dynamics on different timescales and for evaluating their representation in climate models.