Abstract
The structure of thunderstorms has been studied for a long time. In the absence of radar coverage, only high- resolution multifrequency satelliteborne sensors of longer wavelengths (i.e., microwaves) can show structures inside thunderstorms. The National Oceanic and Atmospheric Administration (NOAA) Advanced Microwave Sounding Unit-B (AMSU-B), with five frequencies and 16-km resolution, is now capable of looking at thunderstorm structure. To analyze cloud structure, a tool that can separate regions by size is needed. The à trous wavelet transform, a discrete approximation to the continuous wavelet transform, is such a tool. Images, as well as their wavelet components, may be noisy. To remove noise from wavelet components, those smaller than one standard deviation (of the wavelet image) are equated to zero. This is most suitable for meteorological studies. Images at an appropriate wavelet scale are used for the analysis of thunderstorms. Thunderstorm structures show mostly in scales 2 and 3 (sizes less than 32 and 64 km, respectively) of the à trous transformed images. Other cloud classes are seen either in smaller or larger scales. Given the resolution of the images, three parts of the thunderstorms, namely, the cumulonimbus towers, detraining altostratus, and cirrus anvils, are separated. Thunderstorms in the Indian subcontinent and adjoining seas are grouped according to six classes of wind profiles obtained in this region. Different organizations of towers, altostratus, and cirrus anvils emerged in the AMSU- B images of these six classes.
Corresponding author address: Dileep M. Puranik, Department of Space Sciences, University of Pune, Pune 411007, India. dileepmp@unipune.ernet.in
