TROPICAL AND MONSOON
RESEARCH
Basic Characteristics of the Intraseasonal Variability of the Asian Summer
Monsoon: Diagnostics from ERA and AMIP II Simulation.
The Asian Summer Monsoon (ASM) is the most energetic system of the global circulation in northern summer. Within the monsoon season (JJAS), the ASM exhibits substantial intraseasonal activity, often referred to as active/break cycles, is manifested as fluctuations in convection and circulation that are primarily associated with the westward (10-20 days) and northward propagating (30-60 days) events.
The latent heat release associated with this intraseasonal variability (ISV) can have a large effect on the diabatic heating distribution over the monsoon domain and hence on the global circulation. Although the ASM is recognised as a key element of the global circulation, a comprehensive description of its intraseasonal characteristics in terms of (a) origin and dynamics, (b) 3-dimensional heating and circulation patterns, (c) individual and combined roles in modulating the monsoon activity, (d) relationship with other global circulation features such as Madden-Julian Oscillation (MJO), (e) influence on the extra-tropics and more importantly, (f) relationship to large scale circulation changes due to El Niño Southern Oscillation (ENSO), are not available. The present research addresses these basic characteristics firstly from ERA, observed OLR and CMAP pentad precipitation data, and secondly from the AMIP II run of the Unified Model. The Principal Oscillation Pattern (POP) technique has been applied to understand the origin and propagation features of the ISV. The POP coefficient time series is regressed onto the winds to understand the dynamics of the ISV. The analysis is done for the northern summer monsoon season and for the period 1979-95. The major results are summarised.
During the monsoon season there are two preferred locations of convection, one over the equatorial Indian Ocean and another over the continental region, often referred as a bimodal structure in convection. Our results suggest that this bimodal feature is primarily associated with the 30-60 day time scales and not with the 10-20 day time scales. However, the 30-60 day mode explains more than two thirds of the subseasonal variability in convection and circulation. This mode originates over the Indian Ocean and has complicated propagations (i) eastward over Indian and Pacific Oceans (ii) northward over Bay of Bengal and Indian continent (iii) southward over eastern Indian Ocean, modulating the convection over Australia and (iv) north-westward over the western Pacific. During active phases of the monsoon, the latent heat associated with this mode influences the general circulation. Three dimensional circulation features suggest that the cross-equatorial flow over the monsoon region is the dominant source of moisture. The Rossby waves emanating from the monsoon region propagate across the extra-tropics and circulate over the entire northern hemisphere while the subtropical westerlies act as waveguides. Although the UM captures the gross features of this mode, the propagative features are lacking in their detail. The 10-20 day mode originates over the equatorial central Pacific and propagates west/north-westward, and intensifies over the north west Pacific modulating both the oceanic and continental convection simultaneously. The circulation characteristic of this mode represents a mixed Rossby-Gravity wave. This mode is more regional in structure and the associated latent heating influences only the cross-equatorial flow. As before, the UM fails to capture the propagation characteristics in this time scale also.
The interannual variability of the 30-60 day mode, assessed from the POP coefficient time series, suggests that this mode is strong and coherent in years when the oceanic convection is particularly strong (e.g. 1979, 1986, 1987). Within the period of analysis, all these three years are weak monsoon years. There is no evidence of any systematic relationship with ENSO both in ERA and UM. On the other hand, the 10-20 day mode is strong in years when the continental convection is active (e.g. 1983, 1988, 1994) and all these years are associated with strong monsoon years. Again, no relationship exists between this mode and ENSO. Interestingly, in most of the years the 30-60 day (10-20 day) mode is active during the onset (established) phase of the monsoon. Our results suggest that the characteristics of both modes of the ISV are essentially chaotic and it is the background mean state that is important in predisposing the monsoon system to be strong or weak, consistent with the hypothesis of Charney and Shukla (1981). Our results support those of the theoretical study of Wang and Xie (1997) which suggests that it is the mean circulation that plays an essential role in determining the ISV of the northern summer.
References
Charney. J, and J. Shukla, 1981. Predictability of monsoon. in "Monsoon Dynamics," Cambridge Univ. Press.
Wang, B and X. Xie, 1997: A model for the Boreal Summer Intraseasonal Oscillation. Journ. Atmos. Sci., 54, 72-86 pp.
H. Annamalai and Julia Slingo
CGAM, University of Reading
hanna@met.reading.ac.uk
