The Modulation of ENSO Variability in CCSM3 by Extratropical Rossby Waves

Shayne McGregor, Alex Sen Gupta, Neil J. Holbrook, Scott B. Power
2009 Journal of Climate  
Evidence suggests that the magnitude and frequency of the El Niñ o-Southern Oscillation (ENSO) changes on interdecadal time scales. This is manifest in a distinct shift in ENSO behavior during the late 1970s. This study investigates mechanisms that may force this interdecadal variability and, in particular, on modulations driven by extratropical Rossby waves. Results from oceanic shallow-water models show that the Rossby wave theory can explain small near-zonal changes in equatorial thermocline
more » ... atorial thermocline depth that can alter the amplitude of simulated ENSO events. However, questions remain over whether the same mechanism operates in more complex coupled general circulation models (CGCMs) and what the magnitude of the resulting change would be. Experiments carried out in a state-of-the-art z-coordinate primitive equation model confirm that the Rossby wave mechanism does indeed operate. The effects of these interactions are further investigated using a partial coupling (PC) technique. This allows for the isolation of the role of wind stress-forced oceanic exchanges between the extratropics and the tropics and the subsequent modulation of ENSO variability. It is found that changes in the background state of the equatorial Pacific thermocline depth, induced by a fixed offequatorial wind stress anomaly, can significantly affect the probability of ENSO events occurring. This confirms the results obtained from simpler models and further validates theories that rely on oceanic wave dynamics to generate Pacific Ocean interdecadal variability. This indicates that an improved predictive capability for seasonal-to-interannual ENSO variability could be achieved through a better understanding of extratropical-to-tropical Pacific Ocean transfers and western boundary processes. Furthermore, such an understanding would provide a physical basis to enhance multiyear probabilistic predictions of ENSO indices.
doi:10.1175/2009jcli2922.1 fatcat:fnexsh773rehhhdmegrsvqjmyi