Sensitivity of ENSO to stratification in a recharge-discharge conceptual model

Abstract

El Niño–Southern Oscillation (ENSO) is driven by large-scale ocean–atmosphere interactions in the equatorial Pacific and is sensitive to change in the mean state. Whereas conceptual models of ENSO usually consider the depth of the thermocline to be influential on the stability of ENSO, the observed changes in the depth of the 20ºC isotherm are rather weak, on the order of approximately 5 m over the last decades. Conversely, change in stratification that affects both the intensity and sharpness of the thermocline can be pronounced. Here, the two-strip conceptual model of An and Jin is extended to include three parameters (i.e., the contribution of the first three baroclinic modes) that account for the main characteristics of the mean thermocline vertical structure. A stability analysis of the model is carried out that indicates that the model sustains a lower ENSO mode when the high-order baroclinic modes (M2 and M3) are considered. The sensitivity of the model solution to the coupling efficiency further indicates that, in the weak coupling regime, the model allows for several ocean basin modes at low frequency. The latter can eventually merge into a low-frequency and unstable mode representative of ENSO as the coupling efficiency increases. Also, higher baroclinic modes project more energy onto the ocean dynamics for the same input of wind forcing. Therefore, in this study’s model, a shallower, yet more intense mean thermocline may still sustain a strong (i.e., unstable) and low-frequency ENSO mode. Sensitivity tests to the strength of the two dominant feedbacks (thermocline vs zonal advection) indicate that the presence of high-order baroclinic modes favors the bifurcation from a low-frequency regime to a higher-frequency regime when the zonal advective feedback is enhanced. It is suggested that the proposed formalism can be used to interpret and measure the sensitivity of coupled general circulation models to climate change.