Search

From the literature, we found that PGV-PD3 regressions for on-site earthquake early warning (EEW) can be quite different depending on the presumption whether or not PGV-PD3 data from different regions should be "mixable" in regression analyses. As a result, this becomes a source of epistemic uncertainty in the selection of a PGV-PD3 empirical relationship for on-site EEW. This study is aimed at examining the influence of this epistemic uncertainty on EEW decision-making, and demonstrating it with an example on the use of PGV-PD3 models developed with data from Taiwan, Japan, and Southern California. The analysis shows that using the "global" PGV-PD3 relationship for Southern California would accompany a more conservative EEW decision-making (i.e., early warning is activated more frequently) than using the local empirical model developed with the PGV-PD3 data from Southern California only. However, the influence of this epistemic uncertainty on EEW is not that obvious for the cases of Taiwan and Japan. © 2014 Elsevier Ltd.

From recent lessons, it is evident that earthquake prediction is immature and impractical as of now. Under the circumstances, seismic hazard analysis is considered a more practical approach for earthquake hazard mitigation, by estimating the annual rate of earthquake ground motions (or seismic hazard) based on seismicity and other geological evidences. Like other earthquake studies for the high-seismicity region around Taiwan, this study aims to conduct a new seismic hazard assessment for the region using the well-established FOSM (first-order second-moment) algorithm, on the record of 55,000 earthquakes observed in the past 110 years. The new seismic hazard analysis from a different perspective shows that the annual rate for earthquake-induced PGA to exceed the current design value (i.e., 0.23. g) in two major cities in Taiwan should be relatively low, with it no greater than 0.0006 per year. Besides, the FOSM estimates were found very close to those with Monte Carlo Simulation (MCS), mainly because the skewness of the three random variables (i.e., earthquake magnitude, location, and model error) considered in the probabilistic analysis is not very large.

Earthquake prediction is by all means controversial and challenging, given the fact that some recent catastrophic earthquakes went unpredicted. Not surprisingly, statistical approaches have been utilized to model earthquake randomness in time or space. One of the suggestions is that the earthquake's temporal probability distribution should follow the Poisson model, which is suitable for rare events by definition. As a result, the customarily used hypothesis should be largely associated with the prior judgment that earthquakes are rare, but not as a result of abundant quantitative evidence or theoretical derivation. Therefore, this study aims to offer new empirical evidence to the hypothesis based on 110-year-long earthquake data around Taiwan. From the series of statistical tests, the first statistical inference is indeed in line with the model's proposition: the level of fitting between observation and theory is better for earthquakes with a lower mean rate. To be more specific, it shows that the Poissonian hypothesis applied to local magnitude (ML)3.0 earthquakes around Taiwan with a mean annual rate as high as 1,600 is clearly rejected, but as far as ML7.0 earthquakes with a mean rate of 0.35 per year are concerned, the same hypothesis is statistically accepted for modeling their temporal randomness. Also, according to the tests on a variety of conditions, the annual rate of approximately 0.1 per year (or 10-year return period) was suggested as a reasonable empirical estimate for Poissonian rareness. Accordingly, from a practical point of view, it should be a robust analytical presumption to use the Poisson model in daily earthquake engineering analyses because the return period of design earthquakes is longer than 10 years, if not much longer.

Since the year 1973, more than 54,000 M (w) a parts per thousand yen 3.0 earthquakes have occurred around Taiwan, and their magnitude-frequency relationship was found following with the Gutenberg-Richter recurrence law with b value equal to 0.923 from the least-square calculation. However, using this b value with the McGuire-Arabasz algorithm results in some disagreement between observations and expectations in magnitude probability. This study introduces a simple approach to optimize the b value for better modeling of the magnitude probability, and its effectiveness is demonstrated in this paper. The result shows that the optimal b value can better model the observed magnitude distribution, compared with two customary methods. For example, given magnitude threshold = 5.0 and maximum magnitude = 8.0, the optimal b value of 0.835 is better than 0.923 from the least-square calculation and 0.913 from maximum likelihood estimation for simulating the earthquake's magnitude probability distribution around Taiwan.

Statistical studies on earthquake recurrence time probabilities have frequently been applied to seismic hazard analyses. In Taiwan, an instrumental catalog provides a good opportunity to examine statistical attributes of earthquakes around the study region, with the objective to evaluate the seismic risk and earthquake potential for hazard mitigation. With the completeness of recordings, seismic rates for small-to-moderate magnitudes can be obtained. An elimination of aftershocks was performed using a double-link cluster analysis method. The time intervals between a series of events (the inter-occurrence periods) are stationary with an identical temporal distribution. Based on the goodness-of-fit testing between a few models and observation, we suggested the use of the Gamma distribution in modeling this variable, earthquake inter-occurrence periods, for the study region. Accordingly, unified relationship was constructed, and statistical limitations of sparse sampling for devastating earthquakes (such as M ≥ 6.0 or 7.0) could be resolved. The empirical result evaluated by introducing the conditional probability indicates that the recurrence probability of a M ≥ 7.0 earthquake is 78.8 % within 10 years in Taiwan region. © 2012 Springer Science+Business Media Dordrecht.

Earthquake size can be described with different magnitudes for different purposes. For example, local magnitude M-L is usually adopted to compile an earthquake catalog, and moment magnitude M-w is often prescribed by a ground motion model. Understandably, when inconsistent units are encountered in an earthquake analysis, magnitude conversion needs to be performed beforehand. However, the conversion is not expected at full certainty owing to the model error of empirical relationships. This paper introduces a novel first-order second-moment (FOSM) calculation to estimate the annual rate of earthquake motion (or seismic hazard) on a probabilistic basis, including the consideration of the uncertain magnitude conversion and three other sources of earthquake uncertainties. In addition to the methodology, this novel FOSM application to engineering seismology is demonstrated in this paper with a case study. With a local ground motion model, magnitude conversion relationship and earthquake catalog, the analysis shows that the best-estimate annual rate of peak ground acceleration (PGA) greater than 0.18 g (induced by earthquakes) is 0.002 per year at a site in Taipei, given the uncertainties of magnitude conversion, earthquake size, earthquake location, and motion attenuation.

In this work, two approaches were employed for estimating the spatiotemporal distribution of seismicity density in Taiwan. With the use of the rate-and-state friction model, a model for short-term forecasting according to the fault-interaction-based rate disturbance due to seismicity was considered. Another long-term forecasting model that involves a smoothing Kernel function is proposed. The application of the models to Taiwan led to good agreement between the model forecast and actual observations. Using an integration of the two approaches, the application was found to be capable of providing a seismicity forecast with a higher accuracy and reliability. To check the stability related to the regression the bandwidth function, the forecasted seismicity rates corresponding to the upper and lower bounds of the 95% confidence intervals are compared. The result shows that deviations within the bandwidth functions had an insignificant impact on forecasting reliability. Besides, insignificant differences in the forecasted rate change were obtained when A sigma was assumed to be between 0.1 and 0.4 bars for the application of the rate-and-state friction model. By considering the maximum Coulomb stress change among the seismogenic depth, the model presents a better forecasting ability than that using any single fixed target depth. The proposed methodology, with verified applicability for seismicity forecasts, could be useful for seismic hazard analyses.

This study develops a probabilistic approach for examining site-specific seismic hazards using data from 438 M-w >= 5.5 earthquakes and attenuation relationships local to the region around Taiwan. These are combined to generate semi-observed peak ground accelerations, in a form which can be modeled by a double-lognormal distribution. This model, which satisfies statistical goodness-of-fit, provides the relationship between the exceedance probability and a given ground motion level. The study includes site-specific seismic hazard analysis for four nuclear power plant sites in Taiwan. The results show that the seismic hazards at the four sites are not the same. While no seismic hazard analysis is without challenge, a troublesome trend appears that many applications of decision making are being influenced by the complexity of the calculation, instead of how well the fundamentals of the analysis are understood and can be verified. The study provides an analysis which is not overly complicated, is in good agreement with an empirical control, and offers transparency, traceability, and verifiability. (c) 2012 Elsevier Ltd. All rights reserved.