| Summary: | 碩士 === 國立中央大學 === 太空科學研究所 === 82 === Gravity Wave plays important role in atmospheric dynami- cs,
momentume and energetics. Wind fluctuations observed by MST
radar technique often show complicated temporal behavior .
Waves with different periods may occur at different times.
Conventional FFT analysis is a standard method that provides
the spectrum information of waves. But FT will not allow us to
follow the temporal evolution of the wave spectrum and even
thought STFT (short time Fourier Transform) windowing the
data provides some help in locating time of the even,the time
evolution of wave spectrum can not be descibed satisfa-
ctorily. In this thesis, we introduc the multi-scale,
multi- resolution Wavelet Transform (WT) method to analyze
MST radar velocity data that contains apparent wave-like
struct- ure. Temporal behavior of wave spectra will be clearly
shown in time-frequency plane to help us understand how the
wave spectrum changes. We will discuss the behavior of "
Wavelet Instantaneous Spectrum " (WIS) in each range gate and
study the propagation of Gravity Waves in vertical direction
by the analysis of Wavelet phases . The results obtained by the
WT technique will be compared with the previous studies
carried out by traditional methods:such as FFT, running mean
filtering, cross-correlation method ,etc.. Possible interpr-
etations in terms of nonlinear wave-wave interaction or
wave-mean flow interaction will be proposed. Finally, we
expand the idea of " multi-resolution " to establish the
Wavelet analysis model for Chung-Li radar DBS observation
mode.
|
|---|
