Local and Remote Forcing of the Ocean by the Madden-Julian Oscillation and its Predictability

Abstract

The Madden-Julian Oscillation (MJO) is the dominant mode of intraseasonal variability in the tropical atmosphere and provides global predictability on timescales that bridge the gap between weather and climate. The influence of the MJO on the ocean is explored with a combination of statistical analysis of observations using multivariate time series techniques, dynamical theory, and general circulation models with realistic forcing and bathymetry. The MJO is shown to have a significant and predictable influence on global sea level. Three main regions of influence are identified: (i) the equatorial Pacific and the west coast of the Americas, (ii) the Gulf of of Carpentaria, and (iii) the northeastern Indian Ocean. In the equatorial Pacific, equatorially trapped Kelvin waves are forced by MJO-related surface winds in the western Pacific and propagate eastward. These remotely forced waves then transform into coastal trapped waves that propagate poleward along the west coast of the Americas (consistent with previous work). By way of contrast, in the Gulf of Carpentaria it is shown that the connection with the MJO is due to local wind forcing through simple set-up of sea level. In the northeastern Indian Ocean, a complex sea level pattern involving equatorially trapped Kelvin waves, coastal trapped waves along Sumatra, Java and the Bay of Bengal, and reflected Rossby waves along 5.5$^\circ$N is shown to be caused by a combination of local and remote forcing by MJO-related surface winds.

To examine the predictability of the MJO, and the stability of MJO variability on multidecadal time scales, the MJO index is reconstructed over the last century. The reconstructed index is verified by comparing it with independently observed environmental variables. Three predictability time scales are proposed and estimated from the MJO index. A simple forced damped harmonic oscillator model is used to explain the complex relationship amongst the predictability time scales and also gain insight into the predictability of the MJO.