Important factors for long‐term change in ENSO transitivity

Abstract

AbstractEl Niño and La Niña exhibit significant asymmetry in their duration. El Niño tends to turn rapidly into La Niña after the mature, while La Niña tends to persist for up to 2 years. Reconstructed historical sea surface temperatures (SST) show a significantly increase in the intensity of El Niño‐Southern Oscillation (ENSO) asymmetry, particularly El Niño transitivity, during the last six decades. Atmospheric observational data have shown that the relationship between El Niño and surface zonal wind anomalies over the equatorial Western Pacific (WP) has strengthened, and anomalous WP easterlies have appeared after the 1970s climate regime shift. To investigate the dependency of ENSO transitivity on its amplitude, a suite of idealized experiments using an atmospheric general circulation model (AGCM) is performed by imposing 12 different ENSO‐related SST anomalies exhibiting equal spatiotemporal distribution but different amplitude. Our AGCM experiments show strong nonlinearity in the WP zonal wind against the amplitude of the warm phase. In the strong (weak) El Niño state, the WP response tends to accelerate (prevent) the transition from El Niño to La Niña; however, this relationship is not applicable to the La Niña phase. The asymmetry in transitivity/persistency is found to differ in terms of their sensitivity to the ENSO amplitude; this difference is also detected in the long‐term control simulation of Geophysical Fluid Dynamics Laboratory‐Climate Model version 2.1. The long‐term variation in El Niño transitivity in this model correlates strongly with the decadal variation in the ENSO amplitude and Indo‐Pacific interbasin coupling. However, this result is not apparent in the opposite phase. The results of this study indicate that the decadal change in ENSO amplitude and Indian Ocean feedback could be significantly related to the decadal change in the cyclic nature of ENSO and should be discussed separately for El Niño and La Niña. Copyright © 2012 Royal Meteorological Society