[A-CG33] Multi-scale ocean-atmosphere interaction in the tropics

[ACG33-P04] A new subsurface precursor across the spring predictability barrier for the ENSO prediction

*Zhixiang Zhang^1,2,3,4 (1.Institute of Oceanology, Chinese Academy of Sciences、2.Pilot National Laboratory for Marine Science and Technology Qingdao、3.Center for Ocean Mega-Science, Chinese Academy of Sciences、4.University of Chinese Academy of Sciences)

キーワード：El Niño-Southern Oscillation, Subsurface precursors, Regression analysis, Niño 3.4 index

Despite decades of research on forecasting the El Niño–Southern Oscillation (ENSO) in the past decades, skillfully predicting the ENSO across the spring predictability barrier remains challenging. In this study, we utilize the ensemble of four model products to identify a new subsurface precursor consisting of the anomalous equatorial zonal velocity over 180°–150°W and 140–180 m and potential temperature in the western Pacific (over 120°–140°E and 120–160 m), which could predict the ENSO before spring by applying regression analysis. Such a precursor can predict the Niño 3.4 index in December at a lead time of 13 months with a regressed correlation of 0.79. The anomalous subsurface zonal velocity in the central Pacific favors the eastward migration of the warm water during the development of El Niño. Further validations with diverse training and application periods indicate that for initialization in November, this new precursor shows high skills in predicting the Niño 3.4 index at a lead time of 6–18 months over 1993–2016, outperforming counterparts for the other traditional precursors. Our study can help highlight the importance of subsurface processes in the ENSO development, improving our general ability to predict the ENSO, and enabling better preparedness for the implications of its occurrence.