Recent observations have indicated significant modulation of the Madden–Julian oscillation (MJO) by the phase of the stratospheric quasi-biennial oscillation (QBO) during boreal winter. Composites of the MJO show that upper-tropospheric ice cloud fraction and water vapor anomalies are generally collocated, and that an eastward tilt with height in cloud fraction exists. Through radiative transfer calculations, it is shown that ice clouds have a stronger tropospheric radiative forcing than do water vapor anomalies, highlighting the importance of incorporating upper-tropospheric–lower-stratospheric processes into simple models of the MJO. The coupled troposphere–stratosphere linear model previously developed by the authors is extended by including a mean wind in the stratosphere and a prognostic equation for cirrus clouds, which are forced dynamically and allowed to modulate tropospheric radiative cooling, similar to the effect of tropospheric water vapor in previous formulations. Under these modifications, the model still produces a slow, eastward-propagating mode that resembles the MJO. The sign of zonal mean wind in the stratosphere is shown to control both the upward wave propagation and tropospheric vertical structure of the mode. Under varying stratospheric wind and interactive cirrus cloud radiation, the MJO-like mode has weaker growth rates under stratospheric westerlies than easterlies, consistent with the observed MJO–QBO relationship. These results are directly attributable to an enhanced barotropic mode under QBO easterlies. It is also shown that differential zonal advection of cirrus clouds leads to weaker growth rates under stratospheric westerlies than easterlies. Implications and limitations of the linear theory are discussed.
Recent observations have shown that the strength of the Madden–Julian oscillation (MJO), a global-scale envelope of wind and rain that slowly moves eastward in the tropics and dominates global-weather variations on time scales of around a month, is strongly influenced by the direction of the winds in the lower stratosphere, the layer of the atmosphere that lies above where weather occurs. So far, modeling studies have been unable to reproduce this connection in global climate models. The purpose of this study is to investigate the mechanisms through which the stratosphere can modulate the MJO, by using simple theoretical models. In particular, we point to the role that ice clouds high in the atmosphere play in influencing the MJO.