ebook img

Near-surface characterization from the H/V spectral curves along with the joint inversion of the ... PDF

195 Pages·2017·9.56 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Near-surface characterization from the H/V spectral curves along with the joint inversion of the ...

Universidade de São Paulo Instituto de Astronomia, Geofísica e Ciências Atmosféricas Departamento de Geofísica Irfan Ullah Near-surface characterization from the H/V spectral curves along with the joint inversion of the ellipticity and dispersion curves. SÃO PAULO 2017 I II Irfan Ullah Near-surface characterization from the H/V spectral curves along with the joint inversion of the ellipticity and dispersion curves. Tese apresentada ao Departamento de Geofísica do Instituto de Astronomia, Geofísica e Ciências Atmosféricas da Universidade de São Paulo como requisito parcial para obtenção do título de Doutor em Ciências Área de Concentração: Geofísica Aplicada Orientador: Prof. Dr. Renato Luiz Prado Versao corrigida. O original encontra-se disponivel na unidadae. SÃO PAULO 2017 III Acknowledgement Firstly, I would like to express my sincere gratitude to my advisor Prof. Renato Luiz Prado for the continuous support of my Ph.D. study and related research, for his patience, motivation, and immense knowledge. His guidance helped me in all the time of researching and writing of this thesis. I am really indebted to his kindness. Besides my advisor, I would like to thank my co-advisor and progress report evaluator Prof. Marcello Assumpção for his insightful comments and encouragement and also for the hard question which incited me to widen my research from various perspectives. He also provided me the seismometers for the seismic noise data recording.I am also thankful to Marcello Bianchi for his help in computational aspects. My sincere thanks to the development team and the contributors of the free tool package Geopsy. I would also like to thanks, Dr. Manuel Hobiger and Prof. Francisco J. Sánchez-Sesma for his code sharing and explanation about various aspects via emails. I am also indebted to Luis Galhardo and Felipe Neves for preparing and checking the seismometers prior to the field and their help in the noise recording. I want to thanks to my friends and colleagues Dr. Asif Iqbal, Dr. Taj Ali Khan, Dr. Abdu Raheem, Dr. Hanif Ur Rehman, Aamir Rana, Usama Bilal and Saqib for their moral support. I wish to thanks TWAS-CNPq for the fellowship grant number 190038/2012-8 (CNPq/TWAS - Full-Time Ph.D. Fellowship - GD 2012) and the financial support. The last but not the least, I would like to thank my wife and family who constantly encouraged me to complete the tasks. I will never forget and will always feel my inability for not going to welcome my son (Muhammad Arish Khan) when he was born. IV Abstract The destruction caused by an earthquake at a site depends on many factors like source characteristics such as magnitude, epicentral distance from the site, depth of the source, and on the geological setting of the area. The destruction caused due to the geological setting of an area is termed as site effect. To model the site effect of an area is to determine the shaking level longevity and its displacement amplification. The elastic properties (shear and compressional wave velocities, density, thickness of soil layer, etc.) of the site are required to find out by employing various geophysical procedures. The knowledge of these elastic properties help in better designing the infrastructure, which reduces the chances of destruction caused by a local geological setting due to an earthquake occurrence. This procedure is widely termed as microzonation. The most important parameters for the microzonation are the thickness of soft sediments over the seismic bedrock and its shear wave velocity profile. These two parameters are properly characterized by employing various geophysical techniques like borehole measurement, seismic reflection and seismic refraction. The conventional geophysical methods bring some hindrance to the picture such as, the drilling of a borehole and artificial seismic sources deployment for the reflection and refraction survey, which are both expensive and time consuming, difficult or even in some case impossible to implement in urbanized environment, the investigation is depth limited to few tens of meter. The methods which replaced this conventional geophysical method from the last decades or so is the analysis of Earth vibration caused by the seismic noise which is produced by both natural and cultural sources. This ambient seismic noise can be recorded with less cost and effort with good lateral coverage. Various seismic noise techniques are employed for this job; however, the one which got the most attention in recent years is the horizontal over vertical spectral ratio (H/V) technique. The H/V spectral ratio curve is a fast easy and cheap tool for the near-subsurface characterization. There are various study performed on the topic which has tried to cover almost all the aspects and problems associated with the method. Here in this study, we try to detail the aspects of this technique, which are not been evaluated fully. The different modelling procedures presented to model and physically link the H/V curve with some physical phenomenon will be discussed and its numerical result with the experimental H/V curve will be compared for a borehole test site. The peak and the shape of the H/V curve will be modelled to find its peak frequency deviation from the shear wave resonance frequency by considering different wave-field around the peak. Similarly, the shape dominancy of the H/V curve linkage will be find out. The peak frequency of the H/V curve is used to estimate the thickness-frequency relation by regression analysis. Here we will show that the dispersion curve obtained from multi-channel analysis of surface waves (MASW) can be used to estimate the velocity at one meter and the shear wave velocity increase trend with depth. These values can be used to estimate the thickness frequency relation for an area and its result will be compared with the experimentally derived thickness-frequency relationship for the same area. The sensitivity of the H/V curve shape to the subsurface velocity structure will find out for two main modelling techniques (Rayleigh wave ellipticity and diffused field based H/V curve). The different parts of the H/V curve are inverted (back modelled) to find V out the part of H/V curve which is carrying the most important information about the subsurface structure. The lesson learned from all this analysis will be applied to experimental data of three different sites. The Love waves might contaminate the result of the H/V curve. Two different techniques to remove their effects will be discussed. Then, the joint inversion result of the dispersion and this Love effect removed H/V for more precisely ellipticity curve is discussed. Some new aspects of the H/V curve technique are also discussed at the end. Key Words. ( Microzonation, Earthquake mitigation, Seismic noise, Horizontal-over-Vertical spectral ratio, Modeling of H/V curve, Ellipticity curve, Dispersion curve, Joint inversion of ellipticity and dispersion curve). VI Caracterização da subsuperfície rasa através da curva da razão espectral H/V e da inversão conjunta das curvas de dispersão e de elipticidade Resumo A destruição causada por um terremoto depende de muitos fatores, como características e profundidade da fonte, magnitude, distância epicentral e da configuração geológica da área. A destruição causada devido à configuração geológica da área é denominada como efeito local. A modelagem do efeito local implica na determinação do tempo e nível de vibração e do efeito de amplificação do deslocamento. As propriedades elásticas dos materiais geológicos (velocidade das ondas de compressão e de cisalhamento, densidade, espessura da camada de solo, etc.) podem ser obtidas por diversos métodos geofísicos. O conhecimento dessas propriedades elásticas ajuda a melhor projetar as infraestruturas e reduzir as chances de danos. Este procedimento é denominado de microzoneamento. Os parâmetros mais importantes para realizar o microzoneamento são as espessuras dos sedimentos que recobrem o embasamento e o perfil das velocidades das ondas S (cisalhamento). Esses dois parâmetros são adequadamente caracterizados pelo uso de várias técnicas geofísicas como perfilagens em furos de sondagem, reflexão e refração sísmica. Esses métodos geofísicos trazem algumas restrições como a necessidade da execução de um furo, emprego de fontes sísmicas artificiais que muitas vezes são dispendiosas e por vezes de uso restrito em áreas urbanas, além de muitas vezes estarem limitadas a investigações de apenas algumas dezenas de metros. Os métodos que substituíram esses métodos geofísicos convencionais nas últimas décadas são a análise do ruído sísmico produzido por fontes naturais e culturais. Este ruído sísmico ambiental pode ser registrado com menor custo e esforço e com boa cobertura lateral. Várias técnicas que se utilizam do ruído sísmico podem ser empregadas, no entanto, aquela que obteve maior atenção nos últimos anos é a técnica da razão do espectro horizontal sobre o espectro vertical da onda de superfície (H/V). A curva da razão espectral H/V é uma ferramenta rápida, fácil e de baixo custo para a caracterização da subsuperfície rasa. Existem vários estudos realizados sobre o tema que tentaram cobrir todos os aspectos e problemas associados ao método. Aqui neste estudo são aprofundados alguns aspectos ainda não avaliados em detalhe. Diferentes procedimentos para a modelagem e as associações entre os fenômenos físicos envolvidos e as características da curva H/V são discutidos e os resultados numéricos desses estudos são comparados com informações extraídas de perfis de sondagens de um dos locais estudados. O pico e a forma da curva H / V são modelados para encontrar o desvio na frequência de pico a partir da frequência de ressonância da onda de cisalhamento considerando diferentes campos de onda em torno do pico, assim como sua relação com a forma dominante da curva. A frequência de pico das curvas H/V é utilizada para estimar a relação entre a frequência a espessura através de análise de regressão. O estudo mostra que a curva de dispersão obtida a partir de um ensaio MASW pode ser usada para estimar a velocidade da onda S a um metro de profundidade e sua tendência de aumento com a profundidade. Esses valores podem ser usados para estimar a relação frequência-espessura para uma área. Esses resultados são comparados com a relação frequência-espessura derivada experimentalmente para a mesma área. A sensibilidade da forma da curva H/V à estrutura de velocidade do meio é analisada através de duas técnicas de modelagem (elipticidade da onda Rayleigh e campo difuso baseado na curva H/V). Diferentes partes da curva H/V são invertidas visando avaliar qual a parte da curva H/V contém as informações mais importantes sobre a estrutura subterrânea. As lições aprendidas dessas análises são aplicadas a três dados VII experimentais de locais distintos. As ondas Love podem contaminar o resultado da curva H/V. Duas técnicas diferentes para remover o efeito das ondas amorosas são discutidas. Em seguida, são discutidos os resultados da inversão conjunta das curvas de dispersão e da curva H/V após remoção do efeito da onda Love, ou seja, a curva de elipticidade. Alguns aspectos novos da técnica H/V são discutidos no final. Palavras-chave: microzoneamento, ruído sísmico, razão espectral das componentes horizontais e vertical, curva H/V, curva de elipticidade, curva de dispersão, inversão conjunta das curvas de elipticidade e dispersão VIII Contents Chapter 1: Introduction to the study ............................................................................................................ 1 1. Introduction .............................................................................................................................................. 1 1.1 The objective of the study ...................................................................................................................... 1 1.2 Organization of the thesis ....................................................................................................................... 3 2. Introduction ........................................................................................................................................... 4 2.1 Seismic noise .......................................................................................................................................... 4 2.2. Seismic sources of noise .................................................................................................................... 5 2.3 The composition of noise wavefield ................................................................................................... 6 2.4 Techniques used for seismic noise .................................................................................................... 11 2.4.1 Spatial autocorrelation (SPAC) method ..................................................................................... 11 2.4.2 Frequency-wavenumber analysis ............................................................................................... 14 Chapter 3: Review of the H/V spectral curve method ............................................................................... 18 3.1 Historical overview of the H/V curve ............................................................................................... 18 3.2 Formula of the H/V curve ................................................................................................................ 19 3.3 The experimental procedure for the H/V measurement .................................................................... 25 Chapter 4: Array technique of active source ............................................................................................... 32 4.1 Introduction ....................................................................................................................................... 32 4.2 SASW (Spectral Analysis of Surface Waves) .................................................................................. 33 4.3 MASW (Multichannel Analysis of Surface Waves) ......................................................................... 35 4.3.1 MASW data acquisition procedure ............................................................................................ 35 4.3.2 Length of the geophones spread ................................................................................................. 36 4.3.4 The effect of topography on MASW acquisition ....................................................................... 38 4.4 Dispersion analysis of the acquired data ........................................................................................... 38 4.4.1 Swept-frequency technique ........................................................................................................ 39 4.4.2 Phase-shift method ..................................................................................................................... 42 4.6 Advantage of MASW method........................................................................................................... 44 4.7 Inversion of dispersion curve ........................................................................................................... 45 5.1 The forward and inversion problem .................................................................................................. 48 5.2. Surface wave curves inversion strategies ......................................................................................... 51 5.2.1 Iterative method (Linearized inversion) ..................................................................................... 52 5.2.2 Monte Carlo inversion strategies ............................................................................................... 53 IX 5.2.2.1 Simulated Annealing ............................................................................................................... 53 5.2.2.2 Genetic Algorithm................................................................................................................... 54 5.2.2.3 Neighborhood algorithm ......................................................................................................... 54 5.3 Brief inversion procedure of neighbourhood algorithm .................................................................... 56 5.4. The parameter setting for the inversion ........................................................................................... 58 Chapter 6: Modelling of the H/V curve ...................................................................................................... 60 6. Introduction ......................................................................................................................................... 60 6.1 Nakamura explanation of the H/V curve .......................................................................................... 60 6.2 Rayleigh wave ellipticity modelling of H/V curve ........................................................................... 64 6.3 H/V modelling on the basis of surface wave .................................................................................... 69 6.4 Diffuse field assumption technique ................................................................................................... 73 Chapter7: The peak frequency ( ) and shape of the H/V curve .............................................................. 78 7. Introduction ......................................................................................................................................... 78 7.2 Peak frequency ( ) of the H/V curve .............................................................................................. 79 7.3 Sensitivity analysis of the peak frequency for ellipticity and DFA H/V curves.......................... 83 7.3.1 Thickness (h), density, P-wave and S-wave velocities effects on the peak frequency .............. 86 7.4 The shape of the H/V curve .............................................................................................................. 92 7.5 The numbers of layers and effect on the shape of H/V curve ........................................................... 94 Chapter 8: The use of peak frequency of the H/V curve ............................................................................. 97 8. Thickness frequency relation .............................................................................................................. 97 8.1 How to estimate the value of ................................................................................................ 100 8.2 Thickness frequency relation for the Bebedouro area ..................................................................... 103 8.3 values estimation for the Bebedouro area from MASW dispersion curve ........................... 104 8.4 Deviations in thickness estimates among the three equations ........................................................ 111 Chapter 9: Part of ellipticity and misfit function ...................................................................................... 112 9.1 Which part of ellipticity curve should be used for inversion? ........................................................ 112 9.2 Inversion of the ellipticity curve with velocity trend and thickness information ............................ 116 9.6 Misfit function for the joint inversion ............................................................................................. 123 Chapter 10: The inversion of ellipticity and dispersion curves ................................................................. 129 10.1 Inversion of the experimental data ................................................................................................ 129 10.2 Data acquisition parameters .......................................................................................................... 132 10.3 The retrieval of ellipticity from single 3c-station recording ......................................................... 136 X

Description:
deployment for the reflection and refraction survey, which are both expensive and time consuming, difficult or . ressonância da onda de cisalhamento considerando diferentes campos de onda em torno do pico, protocol for the seismic noise data acquisition, processing and interpretation. A project
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.