In this paper, we attempt to divide the global divergent field at the upper troposphere in contributions from the Hadley, Walker and monsoon circulations, using a monthly mean velocity potential field at 200-hPa level. First, the zonal mean of the velocity potential is analysed to represent the Hadley circulation. The deviation from the zonal mean is then divided into its annual mean and the seasonal cycle parts, which are considered to represent the Walker and monsoon circulations, respectively. The intensities of each circulation are measured by their peaks in the velocity potential field separated in each component. According to this separation, the mean intensities of the Walker, monsoon and Hadley circulations appear to be 120: 60: 40 (× 105 m2 s−1) in January and 120: 90: 45 (×105 m2 s−1) in July, respectively.
Based on this simple definition, interannual variabilities of each circulation are then examined quantitatively using the National Center for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis. The time series of the intensity of the Walker circulation coincides with the Southern Oscillation index (SOI), and the intensity has weakened in recent decades. That of the Hadley circulation indicates intensifying trend in boreal winter.
Finally, the same analysis is applied for the model atmosphere by the Meteorological Research Institute (MRI) coupled atmosphere–ocean general circulation model (CGCM1) with a gradual increase in CO2 at a compound rate of 1% yr−1 for 150 yr. It is shown that the Hadley circulation intensifies by 40% and the monsoon circulation decays by 20% in boreal summer when the global warming has occurred in a century later. The result demonstrates that the proposed simple separation of the tropical circulation in the Walker, monsoon and Hadley components is useful, although it is not rigorous, for the initial assessment of the model response to the global warming.