Observations of Convective Organization by the MJO

As is now well-documented, the convective envelopes that made up the MJO and “MJO-like” events during the CINDY/DYNAMO and TOGA COARE field campaigns were characterized by a wide variety of disturbances. This diversity is not only evident between events, but also within an individual event as it evolves. We investigate the convective makeup of the MJO during the historical record of satellite data going back to 1974. Three independent techniques to assess the scale and type of disturbances within the MJO are applied to the DYNAMO period and compared with results from TOGA COARE and the historical record. One approach uses a tracking algorithm that is particularly well suited to providing a “census” of mesoscale convective systems (MCSs). This can also be adapted to track convectively-coupled equatorial waves (CCEWs) using suitably filtered data. Another metric utilizes a spatio-temporal wavelet transform that can quantify the contribution of CCEWs across an individual MJO envelope. This approach will be discussed in a companion presentation at this conference by Kazu Kikuchi. A third technique measures the variance of various CCEWs within a given MJO event, using brightness temperature filtered for the individual modes. Results reveal that, while the MJO increases the variance of disturbances within its envelope, the distribution of these disturbances across scales and wave type appears to be remarkably uniform regardless of the technique used. To start with, while MCS characteristics vary geographically around the globe, the tracking method strongly suggests that there is nothing special about the MJO in terms of the spatial scale, lifetime, and propagation characteristics of MCSs within its envelope, other than the fact that they occur more often when the MJO is active. A similar result is obtained by measuring the variance of filtered CCEW activity within the MJO envelope versus outside of it. However, although this filtering approach has been and is currently used to assess variations in CCEW activity, we show an example of how it can lead to erroneous results due to the inclusion of “background” variance that is unrelated to the modes being analyzed. Finally, the probability distribution function (pdf) of brightness temperature is analyzed with respect to the MJO. It turns out that while the MJO does modulate the pdf of brightness temperature locally, the pdfs integrated over the tropics or over the entire globe are extremely stable regardless of whether the MJO is active or not. This suggests that the while MJO substantially alters the geographic distribution of convection, it does not necessarily impact the amount of convection on a global scale, perhaps as a result of a global convective-radiative balance on intraseasonal time scales.