AbstractThe conventional Fourier analysis on the sferics at ELF and VLF frequency bands have been explored for decades. Several phenomena, e.g. infrasound and Schumann resonance, have been well studied by the Fourier spectrogram comprehensively. But the core computation of the Fourier analysis is an integration over a window of a specific time length, therefore, the temporal and frequency resolutions are limited not only by the sampling rate but also the length of the integration window. The instantaneous frequency can’t be obtained through this conventional approach precisely.
The Hilbert-Huang transform (HHT) is introduced in this study to analyze the sferics of transient luminous events (TLEs) recorded at the Lulin observatory. The Hilbert–Huang transform decomposes a signal into so-called intrinsic mode functions (IMFs), and derive instantaneous spectrum via the Hilbert transform. The gigantic jets registered by the Taiwan Ground Optical Observation Networks were used in this study and an early signal of frequency-change before the phase of the leading jet is revealed surprisingly. Furthermore this early signal cannot be identified in the conventional Fourier spectrogram. By a simple simulation, this frequency variation is confirmed as a nature of the discharge, not an alias or a false signal generated by the analysis method. The results of the statistics on the TLEs observed from 2012 to 2014 show that this early signal can be discriminated in more than 70% of the gigantic jets, but only less than 30% of other types of the TLEs are able to identify this feature in the HHT spectrogram. This early signal as well as the polarities of the observed gigantic jets strongly agree with the theoretical model proposed by Krehbiel et al. .
|Date of Award||2015|
|Supervisor||Bing-Chih Chen (Supervisor) & Rue-Ron Hsu (Supervisor)|
- Hilbert-Huang transform
- transient luminous events
- gigantic jets