Examining Tropical Cyclone Structure Variability using H*Wind Analyses

The focus in this paper is on the use of H*Wind analyses to examine tropical cyclone structure with an emphasis on the outer wind structure. Understanding changes to outer structure is crucial to accurately forecasting the location and timing of warnings for the onset of gale-force winds, and is a factor in storm motion via beta-effect propagation, as demonstrated by Carr et al. (1997, MWR) with an empirical tangential wind profile. Significant deviations from such a simple wind profile with radius will occur when secondary eyewall formation occurs or when conditions are right for a tropical cyclone to transform into an annular structure. Without a better understanding of tropical cyclone structure variability, accurate wind and surge forecasting for storm damage potential will remain elusive.

A unique set of H*Wind analyses of the surface wind in the Atlantic during the 2003-2005 seasons are examined to evaluate outer wind structure (R34 and R50) changes. Observational data that are fit to the analysis framework include data transmitted from NOAA P3 and G4 research aircraft equipped with the stepped frequency microwave radiometer (SFMR) flown by the Hurricane Research Division (HRD), and the United States Air Force Reserve (AFRES) C-130 reconnaissance aircraft flight-level winds. Additional sources of data include remotely sensed winds from the polar-orbiting satellite platforms. All data are processed to conform to the common height of 10 m and an averaging period of 1-minute maximum sustained wind speed.

An original software program, hereafter referred to as the Tropical Cyclone Structure Analysis Tool or TC-SAT, was written to analyze tropical cyclone structure variability using H*Wind analyses as input. The TC-SAT software was used to analyze 35 tropical cyclones that occurred in the Atlantic and 3 tropical cyclones that occurred in the eastern North Pacific basin from 2003 through 2005. In this research, 564 H*Wind analyses were available of which 508 analyses included dropsonde data, aircraft flight-level reduced data was used in 470 analyses, and 135 analyses included SFMR data. Tropical cyclone structure change was computed over a temporal period of 12 h using an axisymmetric framework.

The Carr et al. empirical wind profile and the Emanuel's (1995a,b, JAS) axisymmetric model imply that an intensifying (weakening) tropical cyclone will be accompanied by an increase (decrease) in outer wind structure. Whereas a small set of H*Wind analyses during the formation stage has the expected increases in R34 values, a considerable percentage of R34 decreases occur during the intensification stages when only increases would be expected. Similarly, a considerable percentage of R34 increases are observed during the decay phase when decreases would be expected. Physical mechanisms for these changes in outer wind structure departures from the simple model will be presented at the Conference.