Interactions between the ITCZ and Convectively Coupled Kelvin Waves

A number of observational studies have shown that Convectively Coupled Kelvin Waves (CCKWs) are strongly affected by the interactions among atmospheric circulation, moisture transport, and deep convection. In particular, observations show that enhanced Kelvin wave activity in the central Pacific Inter Tropical Converegence Zone (ITCZ) is associated with intensified eastward-propagating convective activity. In this presentation, the physical mechanisms of the interactions between the ITCZ and CCKWs are investigated using a simple model based upon the quasi-equilibrium concept, and through analysis of the Outgoing Longwave Radiation (OLR) signal.

Dias and Pauluis (2009, JAS) utilized an idealized model to derive an analytical relation between the speed of CCKWs and the location and width of the ITCZ. This relationship highlights several key features of the modeled CCKWs that are consistent with the observed CCKWs, including weak dispersion and meridional circulation. It is argued that the latter is necessary to have a coherent structure propagating eastwards along the ITCZ, which favors convergence and therefore enhances precipitation within the ITCZ. For an ITCZ width of the order of the equatorial Rossby radius, Kelvin waves propagate at the moist gravity wave speed (about 15 m/s), whereas for a narrow ITCZ, the propagation speed is comparable to the dry gravity wave (about 50 m/s). It is also shown that the CCKWs phase speed increases with increasing ITCZ distance from the equator.

These modeling results suggest a robust dynamical modulation of the ITCZ through CCKWs, and analysis of the OLR signal reveals several aspects of these interdependencies. A wavenumber-frequency spectral analysis of brightness temperature (a proxy for convection) is applied to satellite data in order to filter CCKWs and reanalysis data are used to represent the global atmospheric circulation. In this study, the ITCZ is characterized by a region of low brightness temperature a proxy for both the ITCZ location and width are defined. The phase speed of the CCKW filtered data is determined using the Radon transform method. Linear regression techniques and probability density analysis are applied in order to validate the theoretical predictions. Consistently with the theoretical results, the fastest waves are found when the ITCZ is the furthest from the equator and the narrowest. Conversely, the slowest waves coincide with broad ITCZs that are located near the equator.