A TECHNIQUE FOR DIAGNOSING 3-DIMENSIONAL AGEOSTROPHIC CIRCULATIONS INBAROCLINIC DISTURBANCES ON LIMITED-AREA DOMAINS

The methodology developed by Keyser et al. for representing and diagno sing three-dimensional vertical circulations in baroclinic disturbance s using a two-dimensional vector streamfunction, referred to as the ps i vector, is restricted to f-plane channel-model geometry. The vertica l circulation described by the psi vector consists of the irrotational (or divergent) part of the ageostrophic wind and the vertical velocit y. A key property of the psi vector is that its projections onto arbit rarily oriented orthogonal vertical planes yield independent vertical circulations, allowing separation of a three-dimensional vertical circ ulation into two two-dimensional components, and thus objective assess ment of the extent to which a three-dimensional vertical circulation i s oriented in a preferred direction. Here the methodology for determin ing the psi vector is modified to be suitable for real-data applicatio ns. The modifications consist of reformulating the diagnostic equation s to apply to conformal map projections and to limited-area domains; d espite the desirability of incorporating topography, this task is defe rred to future research. The geostrophic wind is defined in terms of c onstant Coriolis parameter, rendering it nondivergent and thus confini ng the horizontal divergence to the ageostrophic wind. The ageostrophi c wind is partitioned into harmonic, rotational, and divergent compone nts. This three-way partition provides a consistency check of the dive rgent ageostrophic wind recovered from the horizontal divergence field with its counterpart determined from the psi-vector calculation. The modified psi-vector methodology is illustrated for two well-documented East Coast midlatitude cyclones. The first case (the Presidents' Day storm: 1200 UTC 19 February 1979) considers an interpretation that asc ent in the vicinity of a curved upper-level jet-front system may be vi ewed as a superposition of contributions from cross-stream divergent a geostrophic flow associated with a jet streak and from alongstream div ergent ageostrophic flow associated with a baroclinic wave. The second case (the Megalopolitan storm: 1200 UTC 11 February 1983) addresses t he hypothesis of Uccellini and Kocin that vertical circulations transv erse to meridionally displaced upper-tropospheric jet streaks are coup led in a lateral sense. In both of these cases, the diagnoses reveal t hat the cross-stream component of the divergent ageostrophic circulati on isolates meaningful mesoscale signatures coinciding with regions of precipitation and ascent in the vicinity of upper-level jet-front sys tems, whereas the alongstream component is indicative of synoptic scal e vertical motion. Furthermore, it is found that the cross-contour age ostrophic flow, necessary for Lagrangian rates of change of kinetic en ergy in jet entrance and exit regions, is due primarily to the nondive rgent (i.e., harmonic plus rotational) ageostrophic wind. This result suggests that the practice of linking cross-contour ageostrophic winds and vertical motions in jet entrance and exit regions in the qualitat ive assessment of energy transformations in these regions may be probl ematic in the case of upper-level jet-front systems situated in three- dimensional flows.