An attempt using equatorial waves to predict tropical sea surface temperature anomalies associated with the Atlantic zonal mode

Abstract

IntroductionThe forecast for anomalous sea surface temperature (SST) events associated with Atlantic zonal mode, also known as Atlantic Niño/Niña, is full of challenge for both statistical and dynamical prediction models. MethodsThis study combines SST, wind and equatorial wave signal to construct a linear model, aiming to evaluate the potential of equatorial waves in extending the lead time of a skilful prediction for Atlantic Niño/Niña events. Wave-induced geopotential simulated by linear ocean models and potential energy flux calculated using a group-velocity-based wave energy flux scheme are involved to capture the signal of equatorial waves in the model establishment. ResultsThe constructed linear prediction model has demonstrated comparable prediction skill for the SST anomaly to the dynamical models of the North American Multimodel Ensemble (NMME) during the test period (1992-2016). Compared with the statistical forecast using SST persistence, the model notably improves the six-month-lead prediction (Anomaly correlation coefficient increases from 0.07 to 0.28), which owes to the conservation of wave energy in the narrow Atlantic basin that the Rossby waves reflected in the eastern boundary will transfer the energy back to the central equatorial basin and again affect the SST there. ConclusionThis study offers a streamlined model and a straightforward demonstration of leveraging wave energy transfer route for the prediction of Atlantic Niño/Niñas.