Search

Browse

• Communities
• Authors
• Year
• Titles
• Subjects

My Space

• Login

Statistical analysis of seismic sequence data from the island of Zakynthos for the timeperiod 2016-2019

Vasileiadi Michaela-Natalia

Full record

---

Summary

Earthquakes are one of the most dangerous natural disasters that can causewidespread damage and loss of human life. For more than a century the complexgenerating mechanisms and the statistical properties of both foreshockand aftershock sequences have been studied by many scientists. Stochasticmodels, empirical relations, and seismological laws have been proposed toexplain and represent at least some of the observed characteristics of seismicsequences.The aim of this dissertation is the statistical analysis of the seismic sequencebefore the main earthquake (6.6 ML ) of Zakynthos, Greece that tookplace on October 25, 2018 and the aftershock sequence that resulted from it.The data used in this analysis were obtained from the earthquake catalog ofthe Institute of Geodynamics of the National Observatory of Athens (NOA).They refer to the period from 1/06/2016 to 31/05/2019 and extend over anarea which is part of the Hellenic Trench and has intense seismic activity.The relation between earthquake magnitude scales was studied, and anempirical relation between local magnitude ( ML ) and moment magnitude(Mw) was established using weighted linear regression analysis. In order toestablish this relation and due to insufficient data in the seismic catalogs withrecordings on the moment magnitude scale, it was necessary to study a dataset covering a more extended period (2009-2019). The empirical relationwas used to express earthquake magnitudes in the moment magnitude (Mw)scale, and the plots of moment magnitude and cumulative seismic momentevolution over time were created in the R programming environment.Then, the frequency of earthquakes in relation to their magnitude wasstudied (Gutenberg-Richter law), and the b-value parameter was estimated.A prerequisite for the Gutenberg-Richter analysis is the determination of themagnitude of completeness ( Mc ), which was estimated with three differentmethods: (1) by the method of entire magnitude range (EMR), (2) by themethod of maximum curvature (MAXC) and (3) with the Goodness of fit(GFF). Furthermore, the dependence of the b-value parameter on the magnitudeof completeness ( Mc ) and the variation of this parameter over timebefore the main earthquake were examined. The interevent times distributionbetween successive earthquakes was subsequently studied, for both priorand after the main event, by applying 5 different probability distributions(Weibull, Gamma, Pareto, Exponential and Lognormal). The Akaike informationcriterion (AIC) and the Bayesian information criterion (BIC) wereused to compare the distribution models and to select the optimal model.The distribution that provides the best fit for the interevent times beforethe main event is the Weibull distribution, while the one that gives the bestfit for the aftershock sequence is the Pareto distribution.Finally, the frequency of seismic events in the Zakynthos aftershock sequencewas investigated by applying Omori-Utsu’s (O-U) law; the parametersof this law were estimated as p = 1.57, k = 30531.70 and c = 35.79.In addition, the dependence of the O-U parameters (1) on the magnitudeof completeness and (2) on time was studied. The results of the parameters of Zakynthos were compared (1) with the results of the parameters of theaftershock sequence in Hokkaida-Nansei-Oki, Japan (1993), and (2) withthe results of the parameters of the aftershock sequences of the Kefalonia(26/1/2014) and Lefkada (17/11/2015) main events.

Available Files

Services