As part of the National Weather Service (NWS) Modernization and Restructuring Program, WSR-88D (NE-XRAD) Doppler radar installation has been completed at each Weather Service Office in Florida. Recently, this powerful new tool provided unique opportunities for Jacksonville, Tampa Bay, and Melbourne NEXRAD Weather Service Office personnel to investigate tropical cyclone (TC) rainbands for evidence of tornadogenesis. This study provides a radar-based analysis of known tornadic mesocyclones associated with two mature tropical cyclones that were not landfalling in the vicinity of the tornado occurrence, namely, Tropical Storm Gordon (1994) and Hurricane Allison (1995). Based on successful NEXRAD sampling strategies, detailed analyses of storm-scale reflectivity and velocity signatures are conducted in the context of establishing preliminary critical criteria for use in the tornado detection and warning process. Important characteristics were found to include detection of discrete, small diameter Ͼ50 dBZ echos collocated with storm-relative rotational velocities of 6.5-15 m s Ϫ1. Rotational features, although often subtle, were identifiable for an average of 30 min prior to tornado production, with total durations of 1-2 h. Near the time of tornado touchdown, the core diameter of the lowlevel circulation couplets contracted to approximately 1.85 km (1 n mi), leading to an associated increase of shear across the circulation to 0.010 s Ϫ1 or greater. A comparison between the well-studied Great Plains tornadic supercell and the observed TC-tornado cells revealed a common trait of persistence. While the average depth of rotation associated with the TC-tornado cells (3.5 km) was much more shallow than their midwest counterparts, the ratio of depth of rotation to storm top were comparable. However, the shallow depth and weaker detectable rotation of the TC (tornadic) mesocyclones greatly reduced the detection capability of the current WSR-88D mesocyclone algorithm when compared to identification of traditional supercells. Based upon the analyzed data, the authors offer several recommendations to assist operational radar meteorologists with the challenging task of detecting outer rainband tornadoes. Additionally, the authors propose a new WSR-88D scan strategy (volume coverage pattern, VCP) that would provide additional low-level slices in lieu of several current upper-elevation angles. This new VCP would facilitate improved vertical sampling at lower heights where TC mesoscale circulations are most likely to be detected.

The interaction of marginal Tropical Storm Danny (1985) with an upper-tropospheric positive potential vorticity anomaly was examined. The intensification mechanism proposed earlier for mature Hurricane Elena appears to be valid for Danny as well, despite significant differences in the synoptic-scale environment and in the stage of the tropical cyclone prior to the interaction. Both storms experienced rapid pressure falls as a relatively smallscale positive upper potential vorticity anomaly began to superpose with the low-level tropical cyclone center. The interaction is described in terms of a complex interplay between vertical wind shear, diabatic heating, and mutual advection among vortices at and below the level of the outflow anticyclone. Despite this complexity, the superposition principle appears to be conceptually useful to describe the intensification of tropical cyclones during such interactions.

No abstract

No abstract