| description abstract | The question of whether anticyclones can split and break up is readdressed using a numerical, multilayer, primitive equation model. Applying the conservation of integrated angular momentum (IAM) to barotropic and baroclinic vortices, it has been argued that anticyclones can never split, no matter what their structure is. When an anticyclone splits, the IAM has to increase as the newly formed eddies are pushed away from their original center. Conservation of IAM prohibits such an increase. Several numerical simulations, however, have shown anticyclonic splitting. In a multilayer model, a vertical transport of IAM is possible. For counterrotating eddies (an anticyclone on top of a cyclone) it is easy to see that a vertical exchange of IAM allows the eddy to break up. For a compensated or weakly corotating eddy, breakup is only possible when, in addition to a vertical transport of IAM, in the deep layer(s) IAM is exchanged between the core of the vortex and the surrounding fluid. In the presence of a tilting interface, the pressure gradient associated with the sea surface height (SSH) anomaly, in particular its non-equivalent-barotropic part, drives the required exchanges. The non-equivalent-barotropic SSH anomaly is associated with the vertical phase lag of the most unstable eigenmode (m = 2), which develops when this mode gains energy by baroclinic energy conversion. The previous conclusion that anticyclones cannot split on their own should be revised to the following: anticyclones cannot split by barotropic processes alone?baroclinic instability is a necessary ingredient for splitting to occur. | |
