Answers are grounded in this paper and cite their sources.
Assimilating the sea level data from tropical Pacific tide gauges greatly improved the Lamont model prediction of the 1997/98 El Niño while retaining its overall performance during the past few decades. The implication is that the Lamont forecast system is limited by the observational data used for initialization probably probably as much as by its simplified model physics. The sea level measurements in the vicinity of the equator are extremely effective in correcting the model ocean state and preconditioning it for ENSO prediction.
No takes yet. Share an insight, caveat, or question.
Chen et al. (1998) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: