The location of the Inter-Tropical Convergence Zone (ITCZ) plays an important role in the climatology and habitability of the tropical regions. Yet, despite its importance, the basic physics that determine the location of the ITCZ are not fully understood. Observational associations are somewhat ambiguous. Over some parts of the ocean, the equatorial trough, warmest sea-surface temperature (SST) and surface wind convergence tend to coincide. Yet in other regions the warmest SST and equatorial trough coincide but maximum convection occurs nearer to the equator. Earlier studies (e.g., Tomas and Webster 1997) have shown that the discriminating difference is the existence of a strong cross-equatorial pressure gradient (CEPG). Tomas and Webster argue that if a strong enough CEPG exists that the system is inertially unstable and that secondary ameliorating circulations will drive strong off-equatorial convection where the location is determined more by dynamics than lower boundary temperature.

We reexamine the observational record to reestablish the relationships between SST and CEPG using the newer and longer reanalysis data sets of NCEP/NCAR and ECMWF. Noting that Tomas and Webster examined the problem from a parcel method perspective, we use a zonally symmetric version of the NCAR WRF model to examine in detail the consequences of cross-equatorial advection of potential vorticity which occurs when the CEPG is sufficiently large. It is found that for a given CEPG formed by the large scale SST distribution that convection occurs in locations similar to observations. Further experiments are run in which the same magnitude SST is placed at the equator so that the CEPG=0. In this case it is found that the convection is weaker even though similar pressure gradients exist on either side of the equator. Finally, the zonal symmetry of the model is relaxed in order to test the hypothesis that the relaxation of the instability is important in the generation of easterly waves in the tropical system.