Collaboration of intraseasonal oscillation and synoptic-scale disturbances in the South China Sea summer monsoon onset

Previous studies have suggested that the South China Sea summer monsoon (SCSSM) onset is concurrent with the arrival of a 30-60-day northward propagating trough. On the other hand, from a synoptic viewpoint, some studies pointed out that the arrival of a mid-latitude front may be the triggering mechanism of the SCSSM onset. This study attempts to link these two viewpoints and to investigate how theses two very different time-scale oscillations collaborate with each other to induce the onset.

The onset of 1998 SCSSM was examined in the context of the 30-60- and 2-7-day modes. The European Centre for Medium Range Weather Forecasts ERA-40 reanalysis data are subjected to these two ranges of bandpasss-filter. It is found that a Madden and Julian Oscillation (MJO) episode developed over the equatorial Indian Ocean in early May. Prior to the onset, anomalous 30-60-day easterly winds were induced over the South China Sea (SCS) and the western North Pacific (WNP) by the Kelvin wave response ahead of the convection centre of the MJO. The 30-60-day upper-level convergence also dominated over the central SCS. These unfavorable conditions over the SCS made the 2-7-day synoptic-scale disturbances very difficult to induce the onset in this period. After the MJO had moved to the Maritime Continent and the equatorial WNP, it induced a northward-propagating trough over the WNP. Associated with this 30-60-day trough were low-level westerly winds and upper-level divergence over the central SCS and WNP. As the large-scale condition had changed, a cold front on 23 May could then easily trigger the SCSSM onset.

Numerical experiments using a regional climate model were carried out to further ascertain the findings in the case study. In the control experiment with no moving heat source, it is found that the regional climate model can simulate the cold air intrusion events reasonably well. However, the simulation of the eastward-moving MJO was not satisfactory. The timing of the SCSSM onset is two pentads earlier than the observed as well. In the experiment in which a moving heat source is imposed, the timing of the onset is delayed and is much closer to observation. An examination of the model simulation shows that prior to the onset, the moving heat source induced easterly winds and sinking motion over the SCS. Similar to the observations, these conditions pose a very unfavorable environment for the synoptic-scale disturbances over south China to trigger the onset. As a result, the onset is delayed. When the moving heat source has moved to the Maritime Continent, it induces a northward propagating trough over the Bay of Bengal, Indochina, and the SCS. Associated with this northward-propagating trough are low-level westerly winds and upper-level divergence. These provide a favorable environment; hence enhance the synoptic-scale disturbance on 25 May and therefore trigger the SCSSM onset. The main results are consistent with the case study.