| Popis: |
The El Niño Southern Oscillation (ENSO) has profound implications for weather patterns across the globe, and yet there is no consensus on its response to global warming. Several modelling studies suggest a stronger ENSO in global warming scenarios, while other studies find ENSO weakening. Using a broad range of models of the Coupled Model Intercomparison Project phase 6 (CMIP6) and four different types of warming experiments, here we show that the majority of the models predict a stronger ENSO by century-end in Shared Social Pathway (SSP) experiments, and in idealized 1pctCO2 and abrupt 4xCO2 experiments. Several models, however, predict no change or ENSO weakening, especially in the idealized experiments. For most models the strongest forcing (abrupt-4xCO2) is not associated with the strongest ENSO response, while differences between the models are much greater than between warming scenarios. For the long-term (1000 years) response the models disagree even on the sign of change. Furthermore, changes in ENSO SST variability appear to be poorly correlated with the tropical mean state change, while changes in ENSO rainfall variability correlate well with changes in the mean state and, especially, ENSO SST variability. Evaluating changes in the Bjerknes Stability Index for a subset of models we find it to be a poor predictor for ENSO strengthening as this index typically suggests greater stability with warming. We hypothesize that changes to ENSO stability are offset by increases in atmospheric noise or/and potential nonlinear effects. Thus, a robust inter-model mechanism is still lacking to explain a stronger ENSO simulated with global warming, and caution should be exercised when considering ENSO changes based on a single model or warming scenario. |
