Spatial and temporal climatology of atmospheric wave packets

Previous studies have shown that high impact weather events are often associated with Rossby wave packets and model errors can move downstream within a wave packet. It is therefore in the interest of operational forecasters to have a working knowledge of wave packets and their properties, particularly when forecasting in the medium (3-7 day) range. There have been automated cyclone climatologies and model verifications of cyclone forecasts, but there has been limited work on tracking wave packets in a comprehensive way. This study will produce the first climatology using an object-based tracking algorithm applicable to both reanalysis and model forecasts. This new information will be used to better understand the life cycle of a wave packet, from genesis to dissipation and the role wave packets play in cyclone predictability.

Two methods for tracking the magnitude of the envelope of a wave packet – a complex demodulation algorithm, first proposed by Lee and Held (1993), and a stream-wise Hilbert transform technique demonstrated by Zimin et al. (2006) - will be compared. The meridional wind at 300 hPa (chosen to preserve the spectral power of synoptic scale disturbances) was chosen as the variable on which to base wave packet amplitude (WPA) calculations. The WPA is then adjusted using a low-pass butterworth filter to identify objects of greater-than-synoptic scale (wave number of 4 or less) and a nearest-neighbor search algorithm is used to locate significant peaks in WPA and track their movement in time. This process is completed first for the NCEP/NCAR and then the ERA-40 Reanalysis data sets to establish the climatology. Favored methods of genesis and decay, common tracks, and variations in intensity and track with changes in the large scale flow regime will be analyzed,

References:

Lee, S., and I.M. Held, 1993: Baroclinic Wave Packets in Models and Observations. J Atmos. Sci., 50, 1413-1428. Zimin, A.V. et al., 2006: Extracting Envelopes of Nonzonally Propagating Wave Packets. Mon. Wea. Rev. 134, 1329-1333