Search In this Thesis
   Search In this Thesis  
العنوان
MULTI-APPROACHES SEISMIC HAZARD
ANALYSES IN EGYPT /
المؤلف
Badreldin, Hazem Youssef Mohamed Youssef.
هيئة الاعداد
باحث / حازم يوسف محمد يوسف بدرالدين
مشرف / أحمد سيد أبوالعطا
مشرف / عبدالعزيز خيرى عبدالعال
مشرف / فابيو رومانيلى
تاريخ النشر
2022.
عدد الصفحات
247 P. :
اللغة
الإنجليزية
الدرجة
الدكتوراه
التخصص
الجيوفيزياء
تاريخ الإجازة
1/1/2022
مكان الإجازة
جامعة عين شمس - كلية العلوم - قسم الجيوفيزياء
الفهرس
Only 14 pages are availabe for public view

from 247

from 247

Abstract

Site-specific seismic hazard is the essential tool for engineers and decision makers, for structural design and developing building code requirements for the critical structure (e.g., dams and nuclear power plants). For this purpose, this study carried out site specific seismic hazard estimation studies for Nuweiba city, using multi-approaches seismic hazard analysis. In order to develop that, estimated seismic site response, a set of non-invasive active and passive seismic source experiments have been applied.
Seismic hazard assessment, using active faults seismic source model, area source model and combination between fault and area source models, is performed for reference bedrock conditions (Vs30=800 m/s). Seismic hazard results at the bedrock conditions are integrated with the soil response, to estimate the site-specific seismic hazard at the surface of the soft sediments. It highly recommended, to carry out site-specific seismic hazard, at a specific target site in regular base whenever new information is available, in order to update the pre-existing building codes and to mitigate the seismic risk for future earthquakes hitting strategic areas.
For this purpose, the current thesis is organized in four main chapters as follows: Chapter one provides an introduction to the present study, study area, and the aims of this work. Nuweiba city is situated at the GoA water coast, where is the most hazardous seismic source in the Middle East and Eastern Mediterranean. The main goal of this study is to carry out seismic hazard analysis at some selected sites around the GoA and site specific seismic hazard estimation studies for Nuweiba city, using multi approaches seismic hazard analysis. A general overview about the previous seismic hazard studies
169
Chapter 5 Summary and Conclusions
in Egypt and Gulf of Aqaba, has been introduced during this chapter. The seismic hazard maps computed for GoA before the occurrence of 1995 Mw = 7.3 represent a striking example of seismic hazard underestimation in Egypt. The PGA estimated for GoA was ranging between 30-40 gals. The highest calculated PGA was computed by Mourabit et al. (2014); and Hassan et al. (2017b) approximately 0.6 g and both of these studies have been used NDSHA approach
Chapter Two summarizes the seismotectonic setting and seismic activity of the DST-GoA region. DST is 1000 km north–south trending continental plate boundary, representing the northern part of the African-Syrian rift system. GoA is a left-lateral strike-slip fault and it is representing the most southern extension of the DST. The GoA is about 180 km long, 30-40 km wide and 1800 m deep at its maximum. GoA has been described, as a succession of three pull-apart basins: the Dakar Deep, the Aragonese Deep and the Eilat Deep, respectively, from the south to the north. Three major left-lateral strike-slip faults have been identified inside the GoA, from North to South, the Eilat fault, the Aragonese fault, and the Arnona fault, striking about N20E.
Several seismological and paleoseismolgical studies, as well as historical documents pointed that, GoA is the most active part around Egypt and the Eastern Mediterranean region and is the source of many destructive historical earthquakes. Cities around the GoA have experienced different levels of damages, as a result of both historical and instrumental earthquakes resulting in a huge number of fatalities. Paleoseismic evidence shows that, Northern GoA fault zone was the source of approximately 15 historical earthquakes, with magnitudes of more than Magnitude 6. Nevertheless, Historical documentation reported two large damaging earthquakes of
170
Chapter 5 Summary and Conclusions
magnitudes " ~ "7 near the northern tip of the GoA, in 1068 and 1212. The 1068 was a destructive earthquake, the Ayla city (i.e. the ancient Aqaba city) at the head of Aqaba was completely destroyed.
The seismicity at the GoA occurs as earthquake swarmes (e.g., 21 January 1983 (Ml = 5.3), August 3, 1993 (Mw = 6.2), November 22, 1995 (Mw = 7.3), June 27, 2015 (Mw = 5.2), and May 16, 2016 (Mw = 4.9)). For at least 160 yr, the Aqaba earthquake is the largest recorded seismic event along the DST, with Mw = 7.3 and ML = 6.2. This sequence began on 22nd November 1995, by the main shock with Mw 7.3 and continued until 11th December 1997. The rupture was about 56 km length and 10 km width, representing complex source (e.g., two sub-events).The earthquake was strongly felt in Egypt, Saudi Arabia, Jourdan and Palestine, and caused 11 fatalities and 47 injured, most of them were in Nuweiba.
Chapter three includes the seismic site response estimation at Nuweiba city. A set of non-invasive active and passive seismic experiments have been carried out, including the HVSR technique applied to the seismic ambient noise recordings and to estimate the resonance frequency of the sediments cover. In this chapter, methodology, filed measurements, data analysis of the HVSR have been discussed in detailed way.
Single-station microtremor measurements were carried out at 40 sites across the city of Nuweiba, in order to assess the fundamental resonance frequencies. 500 m spacing grid was used in this study, which covers the whole study area that is suitable for microzonation studies. A three components velocity meter (Trillium 120s seismometer, connected to a Taurus digitizer) is used and the seismic noise data were recorded at 200 samples per second. The following parameters are considered during the processing: 70 sec for the window length, 40 for the smoothing bandwidth coefficient and 0.2 for the
171
Chapter 5 Summary and Conclusions
tapering ratio. In the area of interest, all the measured soil sites have F0 ranging from 0.5 to 5.2 Hz, while the amplitude A0 of the H/V curve at F0 is ranging from 2 to 9.
In the same chapter, MASW is discussed, as it is used to estimate the 1D shear wave velocity profile and Vs30 of the subsurface, a classically indicator for site classification in seismic building codes. Field data acquisition was carried out during the summer, using digital multichannel seismograph (Nanometrics StrataView), with 24-bit analog-to digital converter and dynamic range of 110 dB. 48 vertical geophones, having a 4.5 Hz natural frequency, which were used for the acquisition of Rayleigh surface waves. The maximum profile length is 94 m and the geophone spacing is 1 m. The offset between the source and the first geophone, at each extremity of the seismic profile, is 5 m. To improve the signal to noise ratio, the seismic data acquired with a hammer of 10 kg, are stacked during acquisition with approximately 10 shots. 17 active surface wave measurements have been performed in the study area.
Data processing is performed by Geopsy software. Vs30 and soil class for each site is estimated and defined. Vs30 is ranging from 207 m/s to 556 m/s in Nuweiba city. In most of the sites, very thin layer approximately of 2 m thickness with Vs around 200 m/s has been observed, which most probably represents the heterogeneous alluvial deposits the surface layer. The second layer is about 15 m, which is very representative in most of the sites.
Seismic bedrock depth (i.e. sediments thickness) is estimated, using joint inversion of the Rayleigh-wave ellipticity from the HVSR and MASW dispersion curve. Time frequency analysis module (HVTFA), implemented in Geopsy software, have been used for the estimation of Rayleigh wave
172
Chapter 5 Summary and Conclusions
ellipticity. The conditional Neighborhood Algorithm has been used for the joint inversion of the Rayleigh-wave ellipticity and dispersion curve.
At the end of chapter two, of this study, a Power-law regression model is used for the fitting between the estimated fundamental resonance frequency from the HVSR (0.5-5.4 Hz) and the bedrock depths (19-162 m) obtained from the joint inversion of the Rayleigh-wave ellipticity and Rayleigh-wave dispersion curve, in order to propose a new empirical relationship, specific to the study area. These empirical relationship can be used to estimate the bedrock depth by using the fundamental resonance frequency.
Chapter Four deals with the assessment of seismic hazard, using the NDSHA and PSHA. It presents a clear description for the two approaches. All steps of NDSH and PSHA were presented in detail, including their applications on the study area. Regional and extended source approximation of the NDSH are used, to estimate the seismic hazard at ten selected sites around the GoA. The main input parameters for the NDSH in this study are the regional seismicity, seismogenic areas, and fault plane solutions and structural models. In this work, the ground motion parameters are computed, using the NDSHA for 10 Hz cut-off frequency at ten sites around the GoA. Four earthquake source models, have been developed for the 1995 earthquake. These models are used in this study for ground motion simulation, at the same ten sites along the GoA, which have been used for the regional seismic hazard analysis.
Three rupture scenarios (e.g., bilateral rupture scenario, unilateral rupture scenario NE-SW and unilateral rupture scenario SW-NE are used in this study, to simulate the earthquake ground motion and to study the potential effects of this source at the selected sites along the GoA. Green functions are produced, by using both the modal summation technique and discrete wavenumber technique. The rupture is a discrete analogue of a Haskell type
173
Chapter 5 Summary and Conclusions
model, propagating at 80 % of the shear wave velocity of the medium. The maximum calculated PGA was observed at Nuweiba city in the three selected rupture scenarios, which ranges from 0.14g to 0.66g. By applying the rupturing style of 1995 Nuweiba earthquake (i.e. unilateral rupture scenario SW-NE), we observed that, the calculated PGA at Nuweiba site ranges from 0.14g-0.63g. Regarding to earthquake source model, Shamir et al. (2003) and Klinger et al. (2000) seismic source models are resulted in higher ground motion intensity than Baer et al. (2008) and Hofstetter et al. (2003) seismic source models. The maximum spectral acceleration corresponds to a 0.4 s spectral period, which corresponds to: 0.5g and 0.9g in the case of Shamir et al. (2003) earthquake source model, 0.35g and 0.65g in the case of Hofstetter et al. (2003) earthquake source model, 1.5g and 2.3g in the case of Klinger et al. (2000) earthquake source model, and 1.6g and 2.7g in case of Baer et al. (2008) earthquake source model for median and median +2σ respectively.
PSHA has been discussed in the same chapter, in which active fault source model has been used for seismic hazard assessment. Three major left-lateral strike-slip faults, from North to South, the Eilat fault, the Aragonese fault, and the Arnona fault. These faults can accommodate future damaging earthquakes, and they have been considered in seismic source modeling during this study. Mmax at each fault segment is estimated, using Thingbaijam et al., 2017 scaling, as it is the more recent and updated one, and more dataset is included. Slip rate values are used for the estimation of the seismic moment. Anderson and Luco (1983) from2 and form3 have been used, for estimating the seismic productivity (a value) for each fault source.
Area source model has been adopted from the North Africa seismic hazard project (NAF) (Poggi et al., 2020). Earthquake occurrence for each seismic zone is modelled, using the double-truncated Gutenberg-Richter (GR)
174
Chapter 5 Summary and Conclusions
relation. Mmax is derived, as the size of the largest observed (Mobs) event plus 0.5 magnitude units. Mmin is assigned for each seismic zone, equal to 4.0. Ground-motion Prediction Equations, which recently recommended by experts and developed, based upon the seismic data obtained from the active tectonic environments similar to those in and surrounding the studied area were weighted and used for the assessment of seismic hazard in the frame of logic tree approach.
Three models have been considered in seismic hazard calculations: fault source model, area source model, and hybrid source model combines the fault and area source models. OpenQuake engine has been used for the seismic hazard computations. For each model, Seismic hazard curve, Uniform Hazard Spectra (UHS), and Seismic hazard maps are computed. Seismic hazard curves show that the return period for 0.4g is 475 years and the maximum mean SA value is recorded at 0.15 s, which is 1.25 g for 10% of exceedance in 50 years for the case of fault source model. Vs30 is used as a site proxy in the GMPEs, to integrate the site effects in seismic hazard calculation at the rock site condition and to estimate the site specific seismic hazard at the surface of the soft soil for the study area.
Recommendations
Based on the present study, some recommendations are listed in the following points:
- Gulf of Aqaba is the most active seismic source around Egypt, much more studies from different disciplines in geology and geophysics are needed.
- Seismic hazard studies should be used to update the existing seismic building codes.
175
Chapter 5 Summary and Conclusions
- Paleo-seismological studies are much needed for the region along the Gulf of Aqaba, to compile the earthquake catalog and estimate the return periods of large earthquakes.
- Site specific seismic hazard should be updated regularly, whenever anew data is available.
- New ground motion predication equations for Egypt should be developed to study the accurate attenuation of the seismic waves at the selected sites.
- More GPS stations should be installed along the Gulf of Aqaba to study the movement along the different fault segments.