Earthquakes are the result of a sudden release of accumulated energy, and the energy is released in a combination of radiated seismic waves, frictional heating of the fault surface, and cracking of rocks. Specifically, seismic waves radiate from the hypocentre to the ground surface, generating ground shakings. Once the strong ground motion is of considerable duration, it will cause severe damage and facilities. In order to understand earthquake source mechanism and seismic wave propagation, earthquake strong motion recordings become significantly important, and earthquake strong motion networks for collecting strong motion recordings have been implemented in high-seismicity regions, like Mexico, Japan, California, and Taiwan. Moreover, since the first accurate measurement of ground moti...[
Read more ]
Earthquakes are the result of a sudden release of accumulated energy, and the energy is released in a combination of radiated seismic waves, frictional heating of the fault surface, and cracking of rocks. Specifically, seismic waves radiate from the hypocentre to the ground surface, generating ground shakings. Once the strong ground motion is of considerable duration, it will cause severe damage and facilities. In order to understand earthquake source mechanism and seismic wave propagation, earthquake strong motion recordings become significantly important, and earthquake strong motion networks for collecting strong motion recordings have been implemented in high-seismicity regions, like Mexico, Japan, California, and Taiwan. Moreover, since the first accurate measurement of ground motions in 1933, ground motion instruments have been updated from analog-based instruments to fully automatic computer-aided digital recordings, and digitized earthquake recordings are available online from a number of sources.
Located on the collision boundary zone of the Philippine Sea and the Eurasian plates, Taiwan is well-known for the high seismicity. From 1991 to 1996, the Taiwan Strong Motion Instrumentation Program (TSMIP) was implemented to collect more local earthquake data for seismic studies. More than 700 earthquake stations are currently working, forming a dense instrumental network to closely "watch" the regional seismicity. In this study, more than 40000 strong motion recordings were collected through the TSMIP, associated with major earthquakes of local magnitude larger than 5.0 since 1999 to 2013. With an in-house MATLAB program, those strong motion recordings were processed iteratively and consistently, and a strong motion database was established accordingly. Accordingly, a total of 423 outputs associated with each recording were extracted and computed referring to as earthquake general information, amplitude parameters, frequency-related parameters, energy-related parameters, and initial ground motion parameters.
Without direct measurements of the stress and strain on rocks beneath the ground surface, earthquake prediction is almost impossible and still challenging, especially in terms of "When", "Where" and "What size" of a coming earthquake. As a result, earthquake empirical models and statistical analyses play an important role in earthquake studies and engineering applications, and alternatives, such as earthquake early warning and seismic hazard analysis, are proposed as more practical approaches for mitigating seismic hazards.
Specifically, earthquake early warning utilizing initial ground motions to predict coming peak ground motions/earthquake magnitude becomes one alternative for seismic hazard mitigation. Because of natural randomness and uncertainties associated with empirical models between initial motions and peak motions/magnitude, errors can be expected in earthquake early warning. Subsequently, the reliability of the on-site earthquake early warning system in Taiwan is evaluated from an empirical point of view in this study, and the analysis shows that the reliability of the system is around 85%. Moreover, new thresholds for decision-making are proposed to improve the reliability of the system.
On the other hand, with more than 40000 sets of earthquake strong motion parameters available, new empirical models, in regard to the joint distribution between peak ground acceleration (PGA) and cumulative absolute velocity (CAV), and local CAV prediction models, are developed. The joint distribution of PGA and CAV is constructed with the copula approach, and proved capable of successfully capturing the joint dependence of PGA and CAV. Besides, this new bivariate model is applied to reliability assessment and suggest new design values for earthquake-resistant design and. In addition, CAV seismic hazard analysis is proposed in terms of the local CAV prediction models, along with a case study to estimate the annual CAV exceedance rate for Taipei. Moreover, the Bayesian seismic hazard analysis is proposed for providing new return periods for earthquake-resistant design.
Post a Comment