Content uploaded by John A. Scales
Author content
All content in this area was uploaded by John A. Scales
Content may be subject to copyright.
A preview of the PDF is not available
Theory of Seismic Imaging
Abstract and Figures
Seismic imaging methods are currently used to produce images of the Earth's subsurface properties at diverse length scales, from high-resolution, near-surface environmental studies for oil and gas exploration to long-period images of the entire planet. This book presents the physical and mathematical basis of imaging algorithms in the context of controlled-source reflection seismology. The approach taken is motivated by physical optics and theoretical seismology. The theory is constantly put into practice via a graded sequence of computer exercises using the widely available SU (Seismic Unix) software package. The material covered in this book is more than enough for a one-semester graduate course in seismic imaging, by the end of which students will not only be writing their own migration codes, patterned on codes in the book, but will be well-prepared to read the current literature.
Figures - uploaded by John A. Scales
Author content
All figure content in this area was uploaded by John A. Scales
Content may be subject to copyright.
... Amplitudes of reflected arrivals depend on the impedance contrast at the reflector, i.e. the product of seismic velocity and density (e.g. Tatham and McCormack, 1991;Sheriff and Geldart, 1995;Scales, 1997;Yilmaz, 2001). To answer the question why fault reflections are not recorded more often at the strong velocity and density contrast near the AF (sections 4.2.2, 4.2.3), ...
... only P wave propagation (e.g. Scales, 1997). Basically, all studied models consist of two quarterspaces in which velocities vary with depth only. ...
... Therefore, I derived a one-dimensional P velocity model and from that build a traveltime table for all possible offsets and depths in the target volume using a finite-difference (FD) scheme (van Trier and Symes, 1991). The FD scheme is based on the geometrical optics or WKBJ approximation 2 of wave propagation (Scales, 1997). The FD grid spacing is 10 m in horizontal and vertical direction, and traveltimes for distances between grid nodes are interpolated linearily. ...
The Dead Sea Transform (DST) is a prominent shear zone in the Middle East. It separates the Arabian plate from the Sinai microplate and stretches from the Red Sea rift in the south via the Dead Sea to the Taurus-Zagros collision zone in the north. Formed in the Miocene »17 Ma ago and related to the breakup of the Afro-Arabian continent, the DST accommodates the left-lateral movement between the two plates. The study area is located in the Arava Valley between the Dead Sea and the Red Sea, centered across the Arava Fault (AF), which constitutes the major branch of the transform in this region. A set of seismic experiments comprising controlled sources, linear profiles across the fault, and specifically designed receiver arrays reveals the subsurface structure in the vicinity of the AF and of the fault zone itself down to about 3–4 km depth. A tomographically determined seismic P velocity model shows a pronounced velocity contrast near the fault with lower velocities on the western side than east of it. Additionally, S waves from local earthquakes provide an average P -to-S velocity ratio in the study area, and there are indications for a variations across the fault. High-resolution tomographic velocity sections and seismic reflection profiles confirm the surface trace of the AF, and observed features correlate well with fault-related geological observations. Coincident electrical resistivity sections from magnetotelluric measurements across the AF show a conductive layer west of the fault, resistive regions east of it, and a marked contrast near the trace of the AF, which seems to act as an impermeable barrier for fluid flow. The correlation of seismic velocities and electrical resistivities lead to a characterisation of subsurface lithologies from their physical properties. Whereas the western side of the fault is characterised by a layered structure, the eastern side is rather uniform. The vertical boundary between the western and the eastern units seems to be offset to the east of the AF surface trace. A modelling of fault-zone reflected waves indicates that the boundary between low and high velocities is possibly rather sharp but exhibits a rough surface on the length scale a few hundreds of metres. This gives rise to scattering of seismic waves at this boundary. The imaging (migration) method used is based on array beamforming and coherency analysis of P -to-P scattered seismic phases. Careful assessment of the resolution ensures reliable imaging results. The western low velocities correspond to the young sedimentary fill in the Arava Valley, and the high velocities in the east reflect mainly Precambrian igneous rocks. A 7 km long subvertical scattering zone (reflector) is offset about 1 km east of the AF surface trace and can be imaged from 1 km to about 4 km depth. The reflector marks the boundary between two lithological blocks juxtaposed most probably by displacement along the DST. This interpretation as a lithological boundary is supported by the combined seismic and magnetotelluric analysis. The boundary may be a strand of the AF, which is offset from the current, recently active surface trace. The total slip of the DST may be distributed spatially and in time over these two strands and possibly other faults in the area.
... The processing sequence performed in this study included four stages: 1) pre-stack time processing, 2) building an interval velocity model in depth, 3) pre-stack depth migration and 4) post-stack processing ( Fig. 2.3). Migration is a process that moves dipping events to their supposedly true subsurface location and collapses diffraction (Scales, 1995). It can be performed either on CMP gathers prior stacking (pre-stack time migration) or on stacked data (post-stack time migration). ...
... This occurs especially when there is a decrease in velocities with depth for example below a salt layer (such as the Messinian salt layer in the Mediterranean), as average velocities cannot account for such a velocity inversion, the use of interval velocities is essential in order to avoid inaccurate scaling of the time migrated section to depth. The depth migrated section can be considered as close representation of the structural cross-section of the subsurface only if the velocitydepth model is sufficiently accurate (Yilmaz, 2001;Scales, 1995). For example, part of the present research focuses on the Jonah high (Ch. ...
Recent giant gas discoveries within deeply buried structural highs in the middle of the Levant basin have attracted the attention of the industrial and academic communities striving to understand the origin of these structures, their relations with the tectonic history of the basin, and their evolution through time. The location of the Levant basin, at the conjunction of two plate boundaries separating the African Plate from Eurasia in the north and from Arabia in the east, further questions the relationship between the basin’s deformation and the regional plate tectonic processes. In particular, several fundamental questions are addressed here: what is the origin of the deeply buried structures in the basin? Are they related to the Early Mesozoic rifting or to the Late Mesozoic – Early Tertiary Africa-Eurasia convergent phase? How are the structures in the basin related to the closure of the Tethys Ocean, to the onshore Syrian Arc fold belt (~1000 km – long fold belt extending from northern Egypt through Israel to Syria), and the Dead Sea transform? How far west did the Syrian Arc deformation extend, and how it varies in the deep basin? How many deformation episodes occurred, how long was each episode active and are they related to onshore activity?
In order to answer these questions a basin-wide seismic analysis of ~500 2D time-migrated seismic reflection lines with accumulated length of ~27,000 km was carried out. In addition, five regional lines were selected and reprocessed using the Pre-Stack Depth Migration procedure (PSDM) that yields a depth section, correct for geometric distortions, and improves the imaging of the deep part of the sections.
Based on this huge database, horsts were distinguished from folds; geometric relations such as onlap over preexisting structure were identified; thickness variations related to syn-tectonic deposition were examined; and distinct episodes of folding activity were identified. This analysis leads to several new understanding of the basins structure and development that were further tested by gravity and magnetic modeling. In what follows the main findings of this study are described.
Seismic interpretation conducted in this study encompasses the sedimentary section deposited since ~160 Ma, which approximately coincides with the post-rift stage of the basin. Structural maps reveal three regional basement block separated by two structural steps. The continental block is vertically separated from the continental margin block by the 1-2 km high coastal plain step. Farther west (~60 km), the continental margin block is separated from the deep basin block by > 2km high and ~10 km wide Continental Margin Fault Zone (CMFZ).
Superimposed on the regional 3-block structure, the basin hosts a variety of shorter wave structures which are the focus of this study. Interpretation and mapping of 72 folds throughout the basin demonstrated that most of them are asymmetric elongated structures, varying in length from less than 1 km to ~50 km and trending to the NNE with an average azimuth of 360. Few of them are associated with deep reverse faults. Timing of folding activity was identified by examining the geometry of the reflectors approaching each fold and by analyzing thickness variation along them. Combining this analysis with a set of nine isopach maps enabled to determine the temporal and spatial evolution of the folds in the entire Levant basin.
Results show that the main folding activity started during the Santonian mostly along the continental margin block and the CMFZ (more or less the nowadays continental shelf and slope), and also in the deep basin. Then, during the Late-Eocene and Oligocene, folding in the deep basin ceased, while intensive activity continued in the eastern part of the basin (continental margin block and CMFZ). Extensive folding in the deep basin renewed in the Early Miocene and continued during the Middle Miocene. Since the Late Miocene folding activity in the entire basin declined. In the Early Pliocene only few folds were active and only in the eastern part of the basin (the nowadays shelf area), and since the Late Pliocene no folding activity was detected anywhere in the basin
Overall, the structural and the chronological similarity between offshore and onshore (Syrian Arc) folding reveals the existence of a wide deformation zone extending more than 100 km landwards of the present shoreline into Syria, Lebanon, Israel Sinai and western Egypt, and more than 200 km offshore towards the Eratosthenes seamount. This deformation zone started its activity in the Santonian (~85 Ma) and continued till the Early Pliocene (~4 Ma).
In a larger scale deformation in the Levant area lines up with a series of compressional structures observed from Morocco to Syria along the northern margins of the African plate. Hence it is suggested here that the north-east African plate margin which shaped during Early Mesozoic rifting, played an essential boundary condition for the following tectonic evolution. The subsequent Late Mesozoic – Cenozoic compression zone therefore reflect the scissor like closure of the Tethys Ocean, when Africa-Arabia had experienced a counterclockwise rotational northward drift and the thin weak crust of the plate margins deformed accordingly. The interplay of plate kinematics and pre-existing continental weakness zones dictated the deformation orientation in the Levant basin for more than 80 m.y.
Cessation of folding activity in the western part of the basin during Late Eocene – Oligocene is interpreted in this study as a result of the intensive shearing and folding along the CMFZ, which absorbed much of the deformation at that time and “protected” the NW part of the basin. Similarly, declining of folding in the entire basin since the Late Miocene and its final cessation in the Late Pliocene is interpreted as a result of the formation of new plate boundaries near Cyprus and along the Dead Sea Transform, which absorbed much of the deformation.
In addition to the many folds described above, two major high structures were examined, Leviathan and Jonah. The Leviathan high, while seems to be at least partially congruent with the folding activity described above, differ in its dimensions and structure from the majority of the folds in the basin. It portrays a triangular wide shape in contrast to the elongated monocline or anticlines typical to Syrian Arc structures. The analysis presented here agrees with previous studies suggesting that it might be an early Mesozoic large horst, that was reactivated during the contractional phase of the basin and distinct monoclines were formed at its boundaries.
However, unlike the Leviathan high, this study suggests that the Jonah high is an ancient horst that was not reactivated. The Jonah high is particularly enigmatic as it is associated with one of the largest magnetic anomalies in the basin, though no significant gravity anomaly is observed. Previous studies raised several possibilities explaining its origin: an ancient horst related to the early stage of basin formation (Late Paleozoic or early Mesozoic); a Syrian Arc fold (Late Cretaceous to Neogene); a giant volcanic seamount; and an intrusive magmatic body.
This study suggests that the Jonah high is a horst bounded by grabens, most probably formed during continental breakup related to the Neo-Tethys formation. However, unlike other extensional structures that were reactivated and inverted during the Syrian Arc deformation, the Jonah high was never reactivated. Rather, it formed a prominent seamount that persisted for 120-140 million years until the Early Miocene, when it was finally buried. This conclusion was further examined by gravity and magnetic modeling showing that a basement high (horst) can explain the observations without the need to add magmatic intrusions or extrusions.
... Another principle goal of migration is to map the apparent dip that is seen on the zero-offset sections into true dip [2,26,31,33]. True dip angle is always greater than apparent dip angle. Consider a reflector dipping at an angle of θ in the true earth as in Figure 8.5. ...
... ) is known the wave emitted from each point of the source can be calculated, as follows [19]: ...
In this paper, a novel method for in detail detection of the winding axial displacement in power transformers based on UWB imaging is presented. In this method, the radar imaging process is implemented on the power transformer by using an ultra-wide band (UWB) transceiver. The result is a 2-D image of the transformer winding, which is analyzed to determine the occurrence as well as the magnitude of the winding axial displacement. The method is implemented on a transformer model. The experimental results illustrate the effectiveness of the proposed method.
Within seismology, geology, civil and structural engineering, deep learning (DL), especially via generative adversarial networks (GANs), represents an innovative, engaging, and advantageous way to generate reliable synthetic data that represent actual samples’ characteristics, providing a handy data augmentation tool. Indeed, in many practical applications, obtaining a significant number of high-quality information is demanding. Data augmentation is generally based on artificial intelligence (AI) and machine learning data-driven models. The DL GAN-based data augmentation approach for generating synthetic seismic signals revolutionized the current data augmentation paradigm. This study delivers a critical state-of-art review, explaining recent research into AI-based GAN synthetic generation of ground motion signals or seismic events, and also with a comprehensive insight into seismic-related geophysical studies. This study may be relevant, especially for the earth and planetary science, geology and seismology, oil and gas exploration, and on the other hand for assessing the seismic response of buildings and infrastructures, seismic detection tasks, and general structural and civil engineering applications. Furthermore, highlighting the strengths and limitations of the current studies on adversarial learning applied to seismology may help to guide research efforts in the next future toward the most promising directions.
A novel diagnostic processing technique called Conductive Fracture Imaging (CFI) measures hydraulic and conductive fractures using microseismic events as a source. The method was applied to three datasets located in onshore unconventional formations in the United States. CFI results were in all cases first delivered independent of any external diagnostic data and only subsequently compared to multiple diagnostics such as microseismic, fiber cross-well strain (CWS), 3D seismic, and recovered core under supervision of Devon Energy’s Subsurface Team.
The comparison reveals a reasonable agreement of the CFI results with cross-well strain for both height and transverse conductive fracture growth. CFI was able to image fractures out 1 mile from the observation lateral, with fractures imaged in areas of no microseismic activity. Furthermore, CFI successfully quantified the height growth of fractures aligned with the pre-existing faults and how natural structures influence conductivity fracture distribution.
CFI reveals a valid relationship with cored & interpreted conductive, hydraulic, and natural fractures. The method provides dynamic images showing fracture morphology from the near-wellbore into the far-field reservoir. Complimentary analytics of relationships between CFI and reservoir properties, limited entry perforation designs, stress shadowing, and depletion effects may generate significant new observations and key learnings to industry as this technique is more broadly adopted.
In this chapter, we first review the reasons why conventional ultrasonography fails to image the interior of bones. Next we show our recent work on imaging a cortical bone layer with ultrasound. Revealing the shape of the cortex of a bone, in particular its thickness, is of interest for evaluating bone strength. In addition we describe how the process of reconstructing a truthful image of the bone cortex includes the estimation of ultrasound wave-speed in cortical bone tissue. Cortical bone exhibits elastic anisotropy, which causes anisotropy of ultrasound wave-speed as well. Therefore a faithful and high-quality picture of the bone cortex is obtained if wave-speed anisotropy is taken into account during image reconstruction. Capitalizing on prior knowledge on the elastic anisotropy of cortical bone, a procedure for estimating wave-speed and its anisotropy is described. It is based on the measurement of a head-wave velocity and an autofocus approach. The latter relies on the fact that the reconstructed ultrasound image shows optimal quality if the wave-speed model is correct. In order to achieve real-time imaging of a bone cortex, image reconstruction is performed with a delay-and-sum algorithm. Finally, we report recent advances in the measurement of blood flow in cortical bone.
We use machine learning algorithms (artificial neural networks, ANNs) to estimate petrophysical models at seis- mic scale combining well-log information, seismic data and seismic attributes. The resulting petrophysical images are the prior inputs in the process of full-waveform inversion (FWI). We calculate seismic attributes from a stacked reflected 2-D seismic section and then train ANNs to approximate the following petrophysical parameters: P-wave velocity (Vp), density (rho) and volume of clay (Vclay). We extend the use of the Vclay by constraining it with the well lithology and we establish two classes: sands and shales. Consequently, machine learning allows us to build an initial estimate of the earth property model (Vp), which is iteratively refined to produce a synthetic seismogram that matches the observed seismic data. We apply the 1-D Kennett method as a forward modeling tool to create synthetic data with the images of Vp, rho and the thickness of layers (sands or shales) obtained with the ANNs. A nonlinear least-squares inversion algorithm minimizes the residual (or misfit) between observed and synthetic full-waveform data, which improves the Vp resolution. In order to show the advantage of using the ANN velocity model as the initial velocity model for the inversion, we compare the results obtained with the ANNs and two other initial velocity models. One of these alter- native initial velocity models is computed via P-wave impedance, and the other is achieved by velocity semblance analysis: root-mean-square velocity (RMS). The results are in good agreement when we use rho and Vp obtained by ANNs. However, the results are poor and the synthetic data do not match the real acquired data when using the semblance velocity model and the rho from the well log (constant for the entire 2-D section). Nevertheless, the results improve when including rho, the layered structure driven by the Vclay (both obtained with ANNs) and the semblance velocity model. When doing inversion starting with the initial Vp model estimated using the P-wave impedance, there is some gain of the final Vp with respect to the RMS initial Vp. To assess the quality of the inversion of Vp, we use the information for two available wells and compare the final Vp obtained with ANNs and the final Vp computed with the P-wave impedance. This shows the benefit of employing ANNs estimations as prior models during the inversion process to obtain a final Vp that is in agreement with the geology and with the seismic and well-log data. To illustrate the computation of the final velocity model via FWI, we provide an algorithm with the detailed steps and its corresponding GitHub code.
A point-source seismic recording can be decomposed into a set of plane-wave seismograms for arbitrary angles of incidence. Such plane-wave seismograms possess an inherently simple structure that make them amenable to existing inversion methods such as predictive deconvolution. Implementation of plane-wave decomposition (PWD) takes place in the frequency=wavenumber domain under the assumption of radial symmetry. This version of PWD is equivalent to slant stacking if allowance is made for the customary use of linear recording arrays on the surface of a three-dimensional medium. An imaging principle embodying both kinematic as well as dynamic characteristics alows us to perform time migraiton of the plane- wave seismograms. The imaging procedure is implementable as a two-dimensional filter whose independent variables are traveltime and angle of incidence.-Authors
Presents an accurate and computationally efficient method for tracing rays through a medium of heterogeneous velocity. The method was developed as part of a general seismic inversion technique for traveltime information where the medium is discretized into rectangular cells, each of which is characterized by a velocity gradient. However, as a stand-alone tool for forward modeling this method may be applied to other cell geometries and the velocity field need not be a linear function of position in each cell. - from Authors
