ENSO Asymmetry And The Associated Ocean-atmosphere Response

Both observational data and model result indicate that there exists asymmetrybetween El Ni o and La Ni a, and El Ni o tends to be stronger than La Ni a. Theintensities of anomaly in the ocean-atmosphere system between ENSO warm and coldevent are not the same because these two events are of different strength. Therefore, itis necessary to find out the factors that bring about ENSO asymmetry. On the basis ofthe ocean mixed-layer temperature tendency equation, the cause of ENSO asymmetryis studied using Simple Ocean Data Assimilation version2.2.4. The results of bothcase and composite analyses show that the nonlinear term in the equation is animportant cause to ENSO asymmetry. The nonlinear temperature advection term takeseffect during both developing and mature phases of ENSO, and in comparison tomature phase, it plays a more notable role during developing phase. Further, thenonlinear dynamical heating term is divided into three components: zonal, meridionaland vertical components. The analysis results show that zonal and meridionalcomponents tend to give rise to positive SSTA asymmetry in the eastern equatorialPacific, while the vertical component tends to lead to negative asymmetry in the sameregion and its effect is much smaller than the two components mentioned before,therefore, El Ni o is stronger than La Ni a.According to Improved El Ni o Modoki Index(IEMI) and Central PacificIndex/Eastern Pacific Index(CPI/EPI), El Ni o is classified into two different types.The asymmetry between typical El Ni o and the following La Ni a is positive in theeastern equatorial Pacific and negative in the central equatorial Pacific, while theasymmetry between El Ni o Modoki and the following La Ni a has the oppositepattern that is negative asymmetry in the eastern equatorial Pacific and positive in thecentral equatorial Pacific. Hence, the nonlinear rectified effect of ENSO event to thetropical Pacific depends on the type of ENSO. The ocean mixed-layer heat budgetanalysis is also done for the different types of El Ni o and their following La Ni a.The results of composite analysis indicate that there exist different patterns ofnonlinear temperature advection terms corresponding to typical El Ni o and El Ni oModoki, respectively. This term is in favor of the asymmetry between typical ENSOwarm event and the following cold event, while for the asymmetry between El Ni oModoki and the following La Ni a this term tends to have a damping effect.Asymmetry of this type may be more affected by other factors than the nonlineartemperature advection.Because of the different intensities of ENSO warm and cold events, theassociated ocean-atmosphere response will not be the same. The composite analysis isdone for the relevant ocean and atmosphere element fields to obtain the pattern ofocean-atmosphere response. The results show that, corresponding to the positiveSSTA asymmetry in eastern equatorial Pacific, positive precipitation anomaly asymmetry appears in the same region with negative OLR anomaly asymmetry, inwhich region the convection is anomalous more active. In the central equatorialPacific, negative precipitation anomaly asymmetry and positive OLR anomalyasymmetry are in correspondence with negative SSTA asymmetry, and the convectionis suppressed. The resultant zonal temperature gradient is decreased, which leads tothe asymmetry of the low-level wind field appears to be westerly anomaly whichcauses the positive asymmetry of SSHA, and the asymmetry of the high-level windfield appears to be easterly anomaly. Meanwhile, the negative asymmetry of theanomalous wind stress field divergence in eastern equatorial Pacific gives rise to thepositive asymmetry of thermocline depth anomaly. The same analyses are done for thetwo types of El Ni o and the following La Ni a, respectively, the results show twocompletely different asymmetry patterns. Without considering the numeric value, thepattern of asymmetry for typical El Ni o and the following La Ni a is similar to whatmentioned before without classifying different types of El Ni o. While for theasymmetry between El Ni o Modoki and the following La Ni a, its magnitude ofasymmetry is much smaller than the asymmetry of typical events. As for the pattern ofasymmetry, because of the positive SSTA asymmetry in the central equatorial Pacificand the negative asymmetry in the eastern equatorial Pacific, the pattern of asymmetryin the ocean-atmosphere response exhibits significant differences compared to theasymmetry between typical El Ni o and the following La Ni a.