Saçılmış yüzey dalgalarının sismik interferometrisi ile saçıcı konumunun belirlenmesi.
Estimating the location of the scatterer by seismic interferometry of scattered surface waves.
- Tez No: 323746
- Danışmanlar: DOÇ. DR. AYŞE KAŞLILAR
- Tez Türü: Yüksek Lisans
- Konular: Jeofizik Mühendisliği, Geophysics Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2012
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Jeofizik Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 69
Özet
Yüzeye yakın saçıcılar, boşluk, mağara, tünel, maden kuyusu, gömülü atık, arkeolojik kalıntı, su rezervuarları vb yapılardır. Özellikle karstik boşluklar; bina, yol, demiryolu ve benzeri inşaat çalışmaları sırasında veya sonrasında büyük riskler oluşturmaktadır. Bu tür boşluklar rüzgar, güneş ve benzeri geniş alana yayılı güç santralleri için de benzer riskler oluşturmaktadır. Bu nedenle yüzeye yakın saçıcıların belirlenmesi ve izlenmesi olası hasarların en aza indirgenmesi açısından önem taşımaktadır.Bu çalışmada sismik interferometri ile yüzeye yakın saçıcıların konumlarının belirlenmesi amaçlanmıştır. Saçılmış yüzey dalgalarının sismik interferometrisi ile saçıcı konumunun belirlenmesi bu konuda yapılan ilk çalışmadır. Saçılmış yüzey dalgalarının sismik interferometrisi ile elde edilen dalga alanı negatif zamanlarda oluştuğundan hayalet saçılmış yüzey dalgaları (ghost scattered surface waves) olarak isimlendirilmişlerdir. Bu amaç doğrultusunda iki farklı modelleme yöntemi ile yüzeye yakın saçıcıların bulunduğu ortam modellerinden yapay sismogramlar üretilmiştir. İlk modelleme yöntemi, saçılmış dalga alanının integral yöntemiyle modellemektedir. Bu modelleme yönteminin saçılmış dalga alanını direk dalga alanından bağımsız hesaplayabilmesi yöntemin sadece saçılmış dalga alanı üzerinde test edilmesi açısından önemlidir. Bu yöntem ile yapılan modellemelerde, üretilen saçılmış dalga alanına sismik interferometri uygulanarak interferometrik görüntüler elde edilmiştir. Bu görüntülerdeki hayalet saçılmış dalgalardan ters çözüm yardımıyla saçıcının konum parametreleri kestirilmiştir. Ters çözüm için sönümlü tekil değer ayrışımı kullanılmıştır.Diğer modelleme çalışması iki boyutlu sonlu farklar modelleme programı ile yapılmıştır. Bu yöntemle elde edilen sismogramlardaki doğrudan yüzey dalgalarını bastırmak için öncelikle interferometrik yüzey dalgası bastırma işlemi yapılmıştır. Bu işlem sonucunda elde edilen sismogramlar üzerinden benzer şekilde interferometri ve ters çözüm adımları uygulanmıştır.Son olarak, önerilen yöntem ultrasonik laboratuvarda toplanmış bir veri üzerinde denenmiştir. Bu verideki direk yüzey dalgalarını bastırmak için F-K süzgeci kullanılmıştır. Süzgeçlenmiş sismogramlara daha öncekiler gibi interferometri ve ters çözüm adımları uygulanmıştır.Elde edilen sonuçlar, saçılmış dalga alanı üzerinden yapılan interferometri ile elde edilen hayalet saçılmış dalgaları ile saçıcıların konumunun iyi bir şekilde kestirilmesinin mümkün olduğunu göstermiştir.
Özet (Çeviri)
Near-surface structures such as cavities, caves, tunnels, mineshafts, buried objects, archeological ruins, water reservoirs and similar, cause scattered surface waves. These near-surface scatterers may pose risk during and after the construction of buildings, transportation ways (roads, highways, railways) or power plants (wind, solar, etc) which are spread to wide areas. These scatterers can be affected by the changes in the hydraulic regime, earthquakes and change of the loading on the soil and may pose hazards. Therefore, the detection and monitoring of this type of weak zones is important to mitigate environmental and geohazards.Imaging shallow objects by body waves requires high resolution data acquired in a dense spatial array. Surface waves do not require the same dense sampling, since their wavelengths are longer when compared to body waves. Surface waves are widely used in global, exploration and near-surface geophysics. A notable difference in the applications is the frequency content and the array aperture of the measurements which affect the investigation depth. The dispersive property of surface waves allows the estimation of the shear wave velocity structure and attenuation of shallow layers. In global seismology, surface waves are used to investigate the crust and upper-mantle structure (Kovach 1978; Cong & Mitchell 1998; Chang & Baag 2005) and the source properties of seismic events (Canitez & Toksoz 1971; Ekström 2006). In geotechnical engineering shear wave velocity estimation from surface waves has become a popular tool and different techniques are applied to obtain the near-surface properties of the medium. This is relevant for the mitigation of the hazards that may be caused by earthquakes for constructions (Nazarian et al. 1983; Rix et al. 1998; Park et al. 1999; Foti 2000; O?Neill 2003). They are also used in detecting buried objects, which are of great interest in civil and environmental engineering, archeological and land mine explorations. Several authors used scattered surface waves for imaging cavities, buried objects, or shallow water reservoirs (Snieder, 1987; Herman et al, 2000; Campman and Riyanti, 2007; Kaslilar, 2007). The scattered surface waves are studied in detail in terms of seismic interferometry by Halliday and Curtis (2009).This study represents the first time where the seismic interferometry is used to estimate the location of a scatterer using scattered surface waves. Because of the negative arrival times of the scattered surface waves in inteferometric images, they are called ghost scattered surface waves. Seismic interferometry refers to the method of retrieving the interreceiver wavefield by cross-correlating the wavefields recorded at each of the receivers (Halliday and Curtis, 2010; van Manen et al., 2006; Wapenaar, 2004; Wapenaar et al., 2011). Seismic interferometry can be divided into controlled-source interferometry and passive seismic interferometry. Controlled-source interferometry involves cross-correlation along with summation over different source positions, while passive seismic interferometry is the methodology of turning passive seismic measurements into seismic responses. While seismic interferometry was proven useful in constructing surface wave waveforms from passive noise sources (Snieder and Wapenaar, 2010; Halliday and Curtis, 2008), it is also shown that active source signals may also be used to estimate the interreceiver surface waves, which can be used for predictive ground roll removal (Halliday et al., 2007, Halliday and Curtis, 2010).In this study, location of a scatterer is estimated by using seismic interferometry on scattered surface waves. Here, two different modeling methods are used to obtain the scattered wavefields. In the first one, scattered wavefield is calculated with the computationally efficient method developed by Kaslilar (2007). In this method three dimensional propagation and scattering of elastic waves are considered in an isotropic, laterally homogeneous embedding in which bounded objects with contrasting density are present. Since in the method the total wavefield is obtained as the sum of the incident wavefield and the scattered wavefield, , only the scattered part of the wavefield is used in the modeling and in retrieving of the ghost scattered surface waves. Having the chance of using only the scattered part of the wavefield is important for testing the success of the suggested method. With this opportunity, the direct Rayleigh waves, which dominate the interferometric image, are not present in the wavefield and the traveltimes of the interferometric estimate corresponding to ghost scattered surface waves are easily selected.Seismic interferometry is applied to these scattered waves by using only one source and by cross-correlating the reference trace (the trace at the virtual source position) with the rest of the traces, , which are present on the seismic record. Three virtual source positions, at the left, top and right of the scatterer (receivers 1, 26 and 40), is selected for interferometry, which yielded three interferometric images (VS1, VS26 and VS40). Then, the ghost traveltime equation, which is derived for this study, and the ghost scattered surface wave arrival times for each interferometric image are used in inversion. In the inversion, the nonlinear problem is solved iteratively by using damped singular value decomposition (SVD) method. The system of equations for the forward problem is denoted as . In this relation, the difference between the observed (retrieved), and the calculated ghost scattered data is denoted by , the unknown model parameters - the horizontal (x) and vertical (z) locations of the scatterer - are denoted by the vector , while the Jacobian matrix is represented by . The inversion results show that for all three virtual sources, the location of the scatterer is well estimated.In the second part, total wavefields were obtained by using 2D Finite Difference Wavefield Modelling program (Thorbecke, 2011). Seismograms that were produced by this program contained both direct and scattered surface wavefields, but the direct surface waves tend to mask the ghost scattered surface waves in interferometric images. To use the interferometric prediction and subtraction of surface wave method given by Dong et al (2006), a total of 83 shot gathers were obtained by using a line of sources. In this method, an interferometric estimate of several shot gathers are used to obtain a direct Rayleigh wavefield, which is then subtracted from a chosen seismogram with predictive filtering. This suppresses most of the direct surface waves, while keeping the scattered ones. The resulting seismogram is then used in seismic interferometry as described above. Interferometric images from three virtual sources at the left, top and right of the scatterer (receivers 1, 26 and 30) show that despite the effects of direct surface wave remnants and other artifacts, the ghost scattered waves can still be seen clearly. After the inversion process, the location of the scatterer is well estimated for all three virtual sources.In the final part, the proposed method is tested on real data, which was obtained from a ultrasonic laboratory experiment explained in Kaslilar (2007). Similar to the previous numerical modeling, direct Rayleigh waves needs to be removed from the data before the interferometry. However, since there is only one shot gather available, the interferometric prediction of surface waves is not possible. Instead, an F-K filter is used to remove the direct Rayleigh waves. Seismic interferometry is then applied to the filtered seismogram using the procedure described above. Two virtual sources were selected at the top and right of the scatterer (receivers 32 and 45), but due to the high amount of noise in the data, ghost scattered waves are somewhat harder to identify. Here, VS45 contains two separate scattering curves, where one represents the lower left and other represents upper right corners of the scatterer. When estimating the location of the scatterer, these two curves are treated separately. Arithmetic average of the estimated parameters for both curves is considered to represent the scatterer itself.In this thesis, a method for obtaining the location of a near-surface scatterer is proposed by using traveltimes of non-physical (ghost) scattered surface waves retrieved from seismic interferometry. The ghost scattered surface waves are obtained by cross-correlating the recorded scattered surface waves originating from only one source at the surface. The traveltimes of the ghost scattered surface waves are used in the inversion to find the location of the scatterer. The depth and the horizontal position of the scatterer is obtained for different virtual-source locations.Advantage of the proposed method is that the unwanted travel paths between the source and the receiver array are eliminated. These travel paths can traverse a complicated medium. Due to elimination of these paths, the calculation times for waveform inversion studies can be reasonably reduced. Also when lateral changes of the medium properties are present, these path effects can be eliminated by interferometry and locations closer to the target can be considered for estimation of the location of the scatterer.All three cases above demonstrate that the location of the scatterer can be estimated from ghost scattered waves with reasonable accuracy. However, amount of noise in the data can significantly affect the ability of picking the correct travel times for these scattered waves. Therefore, it is observed that this method is more effective on seismic data with good signal-to-noise ratio.
Benzer Tezler
- Saçılmış yüzey dalgalarının çapraz ilişki zamanlarından saçıcı konumunun üç boyutlu kestirimi
Estimating the 3D location of scatterers from cross correlation times of scattered surface waves
AYBARS YİĞİT GARAN
Yüksek Lisans
Türkçe
2015
Jeofizik Mühendisliğiİstanbul Teknik ÜniversitesiJeofizik Mühendisliği Ana Bilim Dalı
PROF. DR. AYŞE KAŞLILAR ŞİŞMAN
- Saçıcıların elastik dalga saçılımı ile görüntülenmesi
Imaging of scatterers through elastic wave scattering
UTKU HARMANKAYA
Doktora
Türkçe
2023
Jeofizik Mühendisliğiİstanbul Teknik ÜniversitesiJeofizik Mühendisliği Ana Bilim Dalı
PROF. DR. AYŞE KAŞLILAR ŞİŞMAN
- Ayrık dalga sayısı yöntemiyle tabakalı ortamda yüksek frekanslı düşey sismik profil sismogramlarının hesaplanması model ve arazi çalışmaları
Başlık çevirisi yok
KENAN GELİŞLİ
Doktora
Türkçe
1992
Jeofizik Mühendisliğiİstanbul ÜniversitesiJeofizik Mühendisliği Ana Bilim Dalı
PROF. DR. ÖMER ALPTEKİN
- İki boyutlu elastik dalga yayınımının sonlu-fark yöntemiyle modellenmesi
The Modelling of the propagation of two dimensional elastic wave by finite-difference method
ZEYNEP KÖMÜRCÜ
Yüksek Lisans
Türkçe
1994
Jeofizik MühendisliğiKaradeniz Teknik ÜniversitesiJeofizik Mühendisliği Ana Bilim Dalı
YRD. DOÇ. DR. VELİ KARA
- Site classification of Turkish national strong-motion recording sites
Türk ulusal kuvvetli yer hareketi kayıt istasyonlarının zemin sınıflandırması
MUSTAFA ABDULLAH SANDIKKAYA
Yüksek Lisans
İngilizce
2008
İnşaat MühendisliğiOrta Doğu Teknik Üniversitesiİnşaat Mühendisliği Ana Bilim Dalı
DOÇ. DR. B. SADIK BAKIR
YRD. DOÇ. DR. M. TOLGA YILMAZ