Wavelet Applications for Mesocyclone Identification in Doppler Radar Observations

Abstract

Severe weather often emanates from thunderstorms that have persistent, well organized, rotating updrafts. These rotating updrafts, which are generally referred to as mesocyclones, appear as couplets of incoming and outgoing radial velocities to a single Doppler radar. Observations of mesocyclones reveal useful information on the kinematics in the vicinity of the storm updraft that, if properly interpreted, can be used to assess the likelihood and intensity of severe weather. Reliable automated detection based on such assessment would serve to enhance the warning process. However, existing data processing methodology in operational mesocyclone detection algorithms is susceptible to noise and natural small-scale variations in Doppler radar data. Reported here is a new wavelet-based approach that offers improved capability for the discrimination of mesocyclones from other shear features. This approach uses discrete wavelets to preprocess the data in order to focus on the known scales where mesocyclones are most likely to reside and extract only the relevant signals for their shape and location identification. Common data quality problems associated with radar observations such as noise and data gaps are also handled effectively as part of the processing. The approach is demonstrated first with a one-dimensional test pattern, then with a two-dimensional synthetic mesocyclone vortex, and finally with a case study.