VORTICITY ASYMMETRIES IN HURRICANE JOSEPHINE (1984)

A further analysis of omega dropwindsonde observations from three syno ptic-flow experiments in the environment of Hurricane Josephine is des cribed, complementing an earlier study by Franklin. The intensity, str ucture and evolution of azimuthal wave-number components of the 850-50 0 hPa layer-mean and deep-layer mean (surface to 100 hPa) vorticity fi elds, including the symmetric component, are investigated, and the det ermination of the extent to which the asymmetries may be interpreted i n terms of the barotropic theory of vortex motion in simple large-scal e environments is sought. The latter endeavour calls for a (necessaril y arbitrary) partitioning of the flow between vortex and environment. As a starting point for this it is hypothesized that the environmental flow ha's a linear variation across a domain that appears to be encom passed by the storm circulation. Based on this assumption it is shown that, on two of the three days on which data were available, the struc ture and strength of the wave-number 1 vorticity asymmetry were simila r to the beta-gyres that are a feature of model calculations. The day- to-day changes in the orientation and strength of these asymmetries in the 850-500 hPa layer-mean calculations are difficult to interpret in the light of barotropic theory, and are possibly associated with vert ical shear in the storm environment, as described in a recent paper by Jones. Such rapid changes are not seen in the corresponding deep-laye r mean calculations, but the wave-number 1 asymmetries in these fields are significantly influenced by the contributions from upper troposph eric levels where there are few data and where the vortex circulation centre is dispatched horizontally far from that at the surface. The or ientation of the wave-number 2 asymmetry varied little over the three- day observation period and was broadly consistent with the orientation of the large-scale deformation field. Moreover, its strength weakened during the period and varied monotonically with that of the deformati on field, a result that is consistent also with theoretical expectatio ns. The attempted analysis highlights certain intrinsic problems in in terpreting hurricane motion in terms of current theories. This is due partly to the non-uniqueness of the partition required to separate the storm environment from the vortex asymmetries and partly because of t he difficulty of determining a representative model-equivalent value f or the environmental absolute-vorticity gradient from the data.