The potential predictability of Singapore and Maritime Continent weather regimes in relation to the MJO and ENSO

The Madden-Julian Oscillation (MJO) is a well-known source of predictability on sub-seasonal-to-seasonal (S2S) time scales and a major driver of intraseasonal weather variability around the globe. For example, the MJO’s interaction with and influence on daily regional weather in the Maritime Continent-Southeast Asia (MC-SEA) region is thought to be most pronounced during boreal winter (November through February), given that the amplitude of MJO activity is often much stronger during that period compared to other times of the year.

In this study, we examine the relationship of the MJO to eight weather regimes (WR) that have been previously defined for Singapore and the MC-SEA region using k-means clustering of daily sounding data from reanalysis. These weather regimes cover the whole annual cycle of rainfall with well-defined peak frequency times and mean spatial structures that correspond to the seasonal movement of the Inter-tropical Convergence Zone (ITCZ) across the Equator. Following previous work, we use a statistical method to compute the lagged relationship between each MJO phase and daily WR occurrence between December 1980 - November 2014 to quantify the change in the likelihood that a certain regime will occur relative to climatology, given an MJO phase in advance. Bimonthly analysis indicates that strong lag relationships exist between MJO phases and certain regimes in different two-month periods, thus giving potential predictability of the type of mean weekly weather in the MC-SEA up to 3-4 weeks ahead. In addition, we consider the modulation of the MJO-WR relationships stratified by the ENSO phase to determine whether the expected WR frequency response to MJO activity varies substantially in different background states.