HR: 0800h
AN: T51A-0315 [Abstracts]
TI: Data Processing and Structure Investigation From Dili Test Shot Experiment Across Central Taiwan
AU: * Chen, H
EM: hwchen@cc.ncu.edu.tw
AF: Dept. of Earth Science and Inst. of Geophysics, National Central University, No. 300,
Jhongda Road., Jhongli, Taoyuan, 32001, Taiwan
AU: Wang, H
EM: apg9005@eq.ccu.edu.tw
AF: Inst. of Seismology, National Chung-Cheng University, No. 168, University Rd., Ming-
Hsiung, Chia-Yi, 621, Taiwan
AU: Yu, Y
EM: bizeasd@yahoo.com.tw
AF: Dept. of Earth Science and Inst. of Geophysics, National Central University, No. 300,
Jhongda Road., Jhongli, Taoyuan, 32001, Taiwan
AB:
A 500 Kg explosive test shot experiment detonated in the town of DiLi with receiver arrays deployed across central
Taiwan was carried out on 1:05 am, 6 October, 2006. The scientific objectives are for preliminary structure
studies, wave propagation effects, efficiency of the energy propagation and attenuation through western
sedimentary basin as well as the data processing strategy for the TAIGER-CD project. More importantly, such a
unique field survey activity provide an important opportunity to verify the strong lateral change of Moho depth
derived from receiver function analysis base on the results of H and Vp/Vs ratio (Łe) study across SSLB and YULB
broadband stations reported by Wang et al., (2006) and the proposed in-active subducting slab in central Taiwan
by Chen et al., (2004).
Overall speaking, the generated energy seems able to reach both end of survey deployment and island-wide as
first arrival travel-time picks are apparent. However, detail data analysis and travel-time inversion studies all show
potential difficulties may encountered in the coming TAIGER active source survey. The identification of latter
phases is highly contaminated by the scattered energy generated from both source signature and near-surface
heterogeneity effects which makes conventional data processing insurmountable. The strong oscillating source
signature makes the conventional spiking and predictive decon ineffective to suppress long tails generated at
shot hole. The nonlinearity of source signal, highly scattered energy, multiples and 3D effects may further mask
later arrivals and obfuscate phase identification. The observed non-impulsive and ringing source signature my
attribute to large borehole diameter, compaction of the explosive and un-evenly distribute booster, PVC borehole
casing and poorly selected site condition. Moho reflections can not be identified from DiLi test shot experiment.
We further implemented the HHT technique of Huang (1998) for signal processing to identify nonlinear and non-
stationary properties of recorded data. The key part is to decompose data into a finite numbers of intrinsic mode
function (IMF) components with time-variable amplitudes and frequencies by using the empirical mode
decomposition (EMD) combined with the Hilbert spectrum analysis (HSA) and power spectral density analysis.
First P- and S- wave arrivals can be identified from decomposed IMFs. The decomposed IMF3 show satisfactory
minimal value of orthogonal index but the significance test indicates that the signals are mainly compose of
noises. Signal-to-noise ratio is low for the recorded Dili test shot data. Study of CMP profile indicates that the
explosive energy decayed within six second with a noticeable but relatively weak signal occurred around 10 sec.
Therefore, the identification of reflected Moho and detachment phases with the characteristic of relative high
frequency content is suspicious.
Travel-time refraction and reflection inversion studies show sedimentary basin across western Taiwan. CMP
profile revealed shallow velocity structure variations but hard to recognize deep reflectors. The maximum depth of
investigation for velocity structure profile obtained is around 25 to 30 km. Careful cooperative plan involves using
one ton explosive on land, safety concerns, field deployment operations, more shot points, geophone group
stacking, timing and detail data processing strategy and preparatory efforts must be carry out immediately.
DE: 0905 Continental structures (8109, 8110)
DE: 8104 Continental margins: convergent
DE: 8109 Continental tectonics: extensional (0905)
DE: 8110 Continental tectonics: general (0905)
SC: Tectonophysics [T]
MN: 2007 Fall Meeting