Similarities between optimal precursors for ENSO events and optimally growing initial errors in El Niño predictions

With the Zebiak–Cane model, the relationship between the optimal precursors (OPR) for triggering the El Nino/Southern Oscillation (ENSO) events and the optimally growing initial errors (OGE) to the uncertainty in El Nino predictions is investigated using an approach based on the conditional nonlinear optimal perturbation. The computed OPR for El Nino events possesses sea surface temperature anomalies (SSTA) dipole over the equatorial central and eastern Pacific, plus positive thermocline depth anomalies in the entire equatorial Pacific. Based on the El Nino events triggered by the obtained OPRs, the OGE which cause the largest prediction errors are computed. It is found that the OPR and OGE share great similarities in terms of localization and spatial structure of the SSTA dipole pattern over the central and eastern Pacific and the relatively uniform thermocline depth anomalies in the equatorial Pacific. The resemblances are possibly caused by the same mechanism of the Bjerknes positive feedback. It implies that if additional observation instruments are deployed to the targeted observations with limited coverage, they should preferentially be deployed in the equatorial central and eastern Pacific, which has been determined as the sensitive area for ENSO prediction, to better detect the early signals for ENSO events and reduce the initial errors so as to improve the forecast skill.