Climate Variation Over The Maritime Continent:Seasonal Prediction And Impact On ENSO And The Asian Monsoon

The Maritime Continent(MC)is characterized by a unique environment where is associated with multiple interactions among weather and climate systems,and it plays an important role in affecting global atmospheric circulation.Using various observational,reanalysis,and hindcast data sets(i.e.NCEP Climate Forecast System version 2(CFSv2)and Project Minerva),this study analyzes the variations of the MC rainfall,their relationship with large-scale climate anomalies,and their seasonalinterannual prediction.The maximum signal-to-noise(MSN)empirical orthogonal function(EOF)analysis is also applied to investigate predictability of low-level atmospheric circulation over the MC and surrounding regions and to identify the most likely disrupters of seasonal predictions.Furthermore,sensitivity experiments are conducted using the Community Earth System Model(CESM)to investigate the influence of the MC on El Ni(?)o-Southern Oscillation(ENSO)evolution.The seasonal evolution of MC rainfall is characterized by a wet season from late winter to early spring and a dry season from late summer to early fall.The MC rainfall is related to the peak-decaying phase of ENSO in the wet season(negatively correlated with the rainfall over East Asia)and to the developing phase of ENSO in the dry season(positively correlated with the rainfall over East Asia).Rainfall over the western Maritime Continent is closely related to the East Asian winter monsoon,and this relationship is independent from ENSO,to some extent.The prediction skill of MC rainfall varies with regions and seasons.It is relatively lower in the wet season(the western MC)than in the dry season(the eastern MC).It is shown in observation that in the wet season the eastern MC rainfall is closely linked to both ENSO and local sea surface temperature(SST)anomalies,while the western MC rainfall is only moderately correlated with ENSO.However,the NCEP CFSv2 hindcast unrealistically predicts the relationships of the western MC rainfall with local SST and ENSO,which contributes to the lower prediction skill for the western MC rainfall.On the other hand,although the model unrealistically predicts the relationship of the eastern MC rainfall with local SST,it well predicts the relationship between the eastern MC rainfall and ENSO,thus leading to a higher skill in the eastern MC.In the dry season,the rainfall over the entire MC is significantly influenced by both ENSO and local SST in observation,and the influences of ENSO and local SST on the MC rainfalls are well captured by the NCEP CFSv2.Therefore,the hindcasts do not show apparently different skill in rainfall prediction for eastern MC and western MC in the dry season.The possible role of atmospheric internal processes is also discussed.The most predictable patterns(MSN EOF1)of atmospheric low-level circulation over the MC and surrounding regions are associated with the developing and maturing phases of ENSO.The other predictable climate processes associated with the low-level atmospheric circulation are more seasonally dependent.For winter and spring,the second most predictable patterns(MSN EOF2)are associated with the ENSO decaying phase.In summer,the second predictable patterns are associated with the western North Pacific monsoon in short leads and with ENSO in longer leads.The second predictable patterns in fall are mainly associated with the tropical Indian Ocean Dipole.The model shows highest prediction skills for the ENSO related predictable patterns,verifying that ENSO is a major skill source for predicting the atmospheric low-level circulation over the MC and surrounding regions.Besides,the intra-ensemble variability shows a strong seasonality that affects different predictable patterns in different seasons.The evolution of intra-ensemble variability reflects the noise-induced differences in the evolution of the predictable signals among ensemble members during summer and fall,and reflects the influences of the extratropical perturbations in the Northern Hemisphere on the seasonal predictability during winter and spring.The CESM model experiments reveal that the existence of MC enhances the nonlinear atmospheric response to the combined seasonal and interannual SST variations(C-mode),which facilitates a quicker termination of El Ni(?)o events.On the other hand,results from the model experiments show that an improved ENSO simulation in the C-mode dynamics is enhanced in the absence of the MC topographical effect(mainly the dynamical lifting effect),which suggests an overestimation of ascending motion over the MC in the model.Since the anomalous western North Pacific anticyclonic circulation serves as a major mediator to bridge ENSO variability and the climate variation over East Asia,the MC impact on anomalous anticyclone evolution further contributes to modification of the relationship between ENSO and the East Asia summer monsoon.Overall,this study provides insight onto improvements for seasonal-interannual prediction of the MC climate variation and its related ocean-atmosphere anomalies.Furthermore,it emphasizes the impact of MC on ENSO decaying phase and demonstrates that a better MC representation may lead to improved prediction of ENSO as well as the post-ENSO Asian summer monsoon.Thus,the study provides a new insight into understanding and predicting ENSO and the Asian monsoon.