Unfolding the differences in two types of ENSO indices with various definitions: A systematic carding with the same criteria

Abstract

El Niño-Southern Oscillation (ENSO) exhibits a wide range of spatial patterns, causing distinct global impacts. Over the past decades, various ENSO indices have been proposed to capture its diversity. Although the classification systems for ENSO flavors differ by definition methodology, it has generally been categorized into two distinct eastern Pacific (EP) and central Pacific (CP) types. However, how different indices characterize these ENSO flavors remains unclear. Here, we reveal the significant differences among these ENSO indices in terms of locally explained variance, type, frequency, and the phase. Specifically, some ENSO indices capture more signals in the equatorial EP or CP, but some others explain relatively insufficient local explanatory variance in the tropical EP and CP, respectively. Moreover, different definition methodologies can result in varying ENSO frequencies identified, even under the same identification criteria and classification process. The most notable discrepancies are observed among CP El Niño and EP La Niña flavors. Furthermore, we discuss the macro-regulation of the Pacific Decadal Oscillation phase on the frequency of each ENSO flavor with different definition. Our findings indicate that each index processes unique characteristics, yet none can comprehensively describe all aspects of ENSO complexity. This highlights the need to consider multiple indices for a comprehensive understanding of ENSO.