Seasonal Synchronization of a Simple Stochastic Dynamical Model Capturing El Niño Diversity

Abstract

Recently, a simple stochastic dynamical model was developed that automatically captures the diversity and intermittency of El Niño–Southern Oscillation (ENSO) in nature, where state-dependent stochastic wind bursts and nonlinear advection of sea surface temperature (SST) are coupled to simple ocean–atmosphere processes that are otherwise deterministic, linear, and stable. In the present article, it is further shown that the model can reproduce qualitatively the ENSO synchronization (or phase locking) to the seasonal cycle in nature. This goal is achieved by incorporating a cloud radiative feedback that is derived naturally from the model’s atmosphere dynamics with no ad hoc assumptions and accounts in simple fashion for the marked seasonal variations of convective activity and cloud cover in the eastern Pacific. In particular, the weak convective response to SSTs in boreal fall favors the eastern Pacific warming that triggers El Niño events while the increased convective activity and cloud cover during the following spring contributes to the shutdown of those events by blocking incoming shortwave solar radiations. In addition to simulating the ENSO diversity with realistic non-Gaussian statistics in different Niño regions, the eastern Pacific moderate and super El Niño and the central Pacific El Niño and La Niña show a realistic chronology with a tendency to peak in boreal winter as well as decreased predictability in spring consistent with the persistence barrier in nature. The incorporation of other possible seasonal feedbacks in the model is also documented for completeness.

A Simple Multiscale Intermediate Coupled Stochastic Model for El Niño Diversity and Complexity

“…First, it has been shown that the climatological SST in the central-eastern Pacific cools in boreal fall and warms in spring as a result of the seasonal motion of the Intertropical Convergence Zone (ITCZ), which also modulates the strength of the upwelling and horizontal advection processes to influence the evolution of the SST anomalies 71 . Since the cool (warm) SSTs tend to coincide with decreased (increased) convective activity and upper cloud cover, a season-dependent damping term, which represents the cloud radiative feedback, is included for describing such a seasonal variation 44 . More specifically, two sinusoidal functions are utilized for parameterizing the otherwise constant α q , which appears as E q = α q T in (6).…”

“…Specifically, WWBs can influence the ENSO development by stimulating eastward-propagating oceanic Kelvin waves, generating surface zonal currents and weakening the evaporation. In fact, many modeling works have been attempted to incorporate semistochastic parameterization for the WWBs 37,[40][41][42][43][44] and suggest that the coupled feedbacks between the interannual SST and the intraseasonal WWBs is sufficient to transfer a damped system to a semi-regular self-sustained oscillator. Likewise, in light of an intermediate coupled model (ICM), Lian et al 45 found that the WWBs are responsible for the existence of the irregularity and intensity of El Niño, whose specific characteristics depend on the timing of the WWBs relative to the phase of the recharge-discharge cycle.…”

A Multiscale Model for El Niño Complexity

“…Specifically, in the centraleastern Pacific, the climatological SST cools in boreal fall and warms in spring as a result of the seasonal motion of the ITCZ, which also modulates the strength of the upwelling and horizontal advection processes to influence the evolution of the SST anomalies 41 . Since the cool (warm) SSTs tend to coincide with decreased (increased) convective activity and upper cloud cover, a season-dependent damping term, which represents the cloud radiative feedback, can account for this seasonal variation in a simple fashion 68 . Besides, the increased wind burst activity in winter as a direct response to the increased atmospheric intraseasonal variability such as the MJO is the main constitution of the seasonal cycle in the western Pacific 61,69 .…”

Conditional Gaussian Systems for Multiscale Nonlinear Stochastic Systems: Prediction, State Estimation and Uncertainty Quantification

“…In particular, the model succeeds in capturing and predicting different super El Niño events [178], including both the directly formed (similar to 1997-1998) and delayed events (similar to 2014-2016). An improved version of the model is also able to recover the season synchronization [179]. The coupled model is as follows: Atmosphere…”