What Controls ENSO‐Amplitude Diversity in Climate Models?

Climate models depict large diversity in the strength of the El Niño/Southern Oscillation (ENSO) (ENSO amplitude). Here we investigate ENSO‐amplitude diversity in the Coupled Model Intercomparison Project Phase 5 (CMIP5) by means of the linear recharge oscillator model, which reduces ENSO dynamics to a two‐dimensional problem in terms of eastern equatorial Pacific sea surface temperature anomalies (T) and equatorial Pacific upper ocean heat content anomalies (h). We find that a large contribution to ENSO‐amplitude diversity originates from stochastic forcing. Further, significant interactions exist between the stochastic forcing and the growth rates of T and h with competing effects on ENSO amplitude. The joint consideration of stochastic forcing and growth rates explains more than 80% of the ENSO‐amplitude variance within CMIP5. Our results can readily explain the lack of correlation between the Bjerknes Stability index, a measure of the growth rate of T, and ENSO amplitude in a multimodel ensemble. Key Points Large contribution to El Niño/Southern Oscillation (ENSO)‐amplitude diversity in climate models originates from stochastic forcing Competing effects in the ENSO dynamics present an important source for ENSO‐amplitude diversity Due to the importance of stochastic forcing and competing processes, the growth rate of sea surface temperature anomalies must not necessarily correlate with ENSO amplitude