The magnitude of El Niņo warming, defined as the SST anomaly in the central and eastern equatorial Pacific, is a key parameter in determining its climatic effect world-wide (Trenberth et al. 1998). What controls the magnitude of El Niņo warming, however, is not well understood. This lack of unnderstanding is highlighted by the concern over whether El Niņo become more energentic in response to global warming. The two most intense El Niņo events of the last century-the 1982-83 El Niņo and the 1997-98 El Niņo occured during the last two decades (Fig.1 ). Is this a manisfitation of the effect of the anthropogentic forcing or the natural variability of the coupled ocean-atmosphere system? Coupled general circulation models (GCMs) are the most sophisticated tools to simulate ENSO and predict its resposne to global warming, but underestimate of the magnitude of El Niņo warming is a general problem of current coupled climate models (Knutson et al. 1997, Meehl and Arblaster 1998) ). For example, the NCAR climate system model, while successful in eliminating the climate drift that had plagued the modeling community for years, underestimate substantially the interannual variability in the equatorial eastern Pacific (Meehl and Arblaster 1998). This not only casts doubt on the predction of the models, but also highlight our lack of understanding of what controls the magnitude of El Niņo warming. Recent paleoclimate findings provide additional urgency for better understanding what controls the magnitude of El Niņo warming (Sandweiss et al. 1996, Rodbell et al 1998). These findings have suggested that there were past climate regimes that had much weaker or even no ENSO.