A stability analysis of ENSO over multi-decadal time-scales in a 1000-yr coupled general cirlculation model simulation
Abstract
The Bjerknes (BJ) stability index is an approximate formula that may be used to analyse the stability of the coupled El Nino/Southern Oscillation (ENSO) mode. The BJ stability index depends on the basic mean state, the atmosphere's sensitivity to sea surface temperature and the ocean's sensitivity to wind forcing. A positive BJ stability index indicates a growth of ENSO stability while a negative BJ stability index signifies the rate of decay with negative contributions coming from damping by mean advection and air-sea heat fluxes while positive contributions from the zonal advective, Ekman pumping and thermocline feedback. Here we employ the BJ stability index to investigate ENSO variability over multi-decadal time scales in the Commonwealth Scientific and Industrial Research Organisation Mark version 3.0L (CSIRO Mk3L) coupled general circulation model and assess the sensitivity of ENSO to changes in tropical climate conditions. Model simulations are forced with atmospheric concentrations fixed at pre-industrial levels and integrated over 1000-years to allow a variety of ENSO events to be simulated. The BJ index points to a simulated ENSO that is damped with the negative contribution from thermal damping by air-sea heat fluxes dominating over the feedback processes. There also exists a significant positive correlation between the ENSO amplitude and ENSO stability such that the BJ index becomes less damped with increasing ENSO variability. We find that significant variations in the ENSO amplitude exist on decadal and centennial time scales. These variations are mainly governed by variations in the strength of the thermocline feedback, which is in turn a function of the coherence between fluctuations of the atmosphere and the ocean in the equatorial Pacific, reflecting the response of winds to temperature gradients and the response of the thermocline to winds. Surprisingly, such decadal and centennial variations are only weakly control by the mean state differences. The implications for interpreting observed ENSO variability will be discussed.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFMOS53A1950B
- Keywords:
-
- 4522 OCEANOGRAPHY: PHYSICAL / ENSO