Improved Velocity Model Research Articles

Synthetic Data Generation for Deep Learning-Based Inversion for Velocity Model Building

Recent years have seen deep learning (DL) architectures being leveraged for learning the nonlinear relationships across the parameters in seismic inversion problems in order to better analyse the subsurface, such as improved velocity model building (VMB). In this study, we focus on deep-learning-based inversion (DLI) for velocity model building, leveraging on a conditional generative adversarial network (PIX2PIX) with ResNet-9 as generator, as well as a comprehensive mathematical methodology for generating samples of multi-stratified heterogeneous velocity models for training the DLI architecture. We demonstrate that the proposed architecture can achieve state-of-the-art performance in reconstructing velocity models using only one seismic shot, thus reducing cost and computational complexity. We also demonstrate that the proposed solution is generalisable across linear multi-layer models, curved or folded structures, structures with salt bodies as well as higher-resolution structures built from geological images through quantitative and qualitative evaluation.

Atlantis — 20 years of seismic innovation finally removes the shroud of mystery

The Atlantis Field has gone through more than two decades of continuous seismic imaging efforts, during which time many innovative technologies were incubated, the most recent one being the successful application of full-waveform inversion (FWI) in salt environments. This technique led to a significant improvement in the subsalt image. However, imaging challenges remain for the Atlantis reservoirs, primarily due to the complex overburden salt geometries and the highly compartmentalized reservoir. Even with an improved velocity model from FWI, the conventional reverse time migration (RTM) images still suffer from illumination issues and contain strong migration swings that hinder the subsalt imaging and subsequent interpretations. Furthermore, early versions of FWI employed an acoustic assumption, leading to visible salt halos at the salt boundaries in the velocity model, which adversely impacted the reservoir imaging. In the last 12 months, elastic time-lag FWI (TLFWI) and FWI-derived reflectivity (FDR) imaging using long-offset ocean-bottom node data have minimized these imaging issues at Atlantis, providing another step change in subsalt understanding. Although the 3D RTM images using the elastic FWI velocity model are similar overall to their acoustic counterparts, the 4D time-lapse RTM images at Atlantis show noticeable improvements. Furthermore, FDR images derived from elastic FWI velocities show obvious benefits over the acoustic ones. With a more accurate modeling engine that allows for better match between synthetic and real data, FDR imaging shows improved illumination, higher signal-to-noise ratio, and better reservoir details over acoustic FDR imaging. This recent advancement in using elastic TLFWI has had immediate positive effects in facilitating the Atlantis Field's current and future development.

A new 1D velocity model and absolute locations image the Mayotte seismo-volcanic region

In May 2018, a seismically quiet region of the Indian Ocean awoke. More than 130 magnitude 4+ earthquakes were recorded in the first month, including a MW 5.9 event on May 15th, 2018. This seismic activity was later identified as being related to an exceptional underwater volcanic eruption offshore Mayotte island, which had emitted more than 6.5 km3 of lava by the time of writing. To better constrain the geodynamic processes responsible for the seismic and volcanic activity, a new network of ocean-bottom seismometers and land stations has been deployed around the seismically active region since February 2019. We present here an improved 1D velocity model for the active area and relocations of manually-picked earthquakes using this new model. The best-constrained events image detailed structures within two clusters of seismic activity east of Mayotte. The westernmost, proximal cluster, close to Mayotte's Petite-Terre island, has a “donut” shape horizontally and an “hourglass” shape in depth. The events distribution suggests the presence of a magma reservoir at around 27 km depth, with earthquakes focused along its sides, and a collapsing system underneath, related to the drainage of another, deeper magma storage zone. The distal cluster, focused 30–50 km offshore of Petite-Terre island, highlights the propagation of a dike between 45 and 25 km depth, aligned towards the new volcanic activity on the seafloor. We interpret this cluster as the fluid pathway towards the new volcano and nearby active seafloor lava fields. The improved velocity model also permits more robust daily monitoring of the seismicity using land stations, allowing local authorities to better assess seismic and volcanic hazards and to communicate them to the island's population.

A reflection-based efficient wavefield inversion

Full-waveform inversion (FWI) is popularly used to obtain a high-resolution subsurface velocity model. However, it requires either a good initial velocity model or low-frequency data to mitigate the cycle-skipping issue. Reflection-waveform inversion (RWI) uses a migration/demigration process to retrieve a background model that can be used as a good initial velocity in FWI. The drawback of conventional RWI is that it requires the use of least-squares migration, which is often computationally expensive, and it is still prone to cycle skipping at far offsets. To improve the computational efficiency and overcome cycle skipping in the original RWI, we have incorporated it into a recently introduced method called efficient wavefield inversion (EWI) by inverting for the Born-scattered wavefield instead of the wavefield itself. In this case, we use perturbation-related secondary sources in the modified source function. Unlike conventional RWI, the perturbations are calculated naturally as part of the calculation of the scattered wavefield in an efficient way. Because the sources in the reflection-based EWI (REWI) are located in the subsurface, we are able to update the background model along the reflection wavepath. In the background velocity inversion, we calculate the background perturbation by a deconvolution process at each frequency. After obtaining the REWI inverted velocity model, a sequential FWI or EWI is needed to obtain a high-resolution model. We determine the validity of our approach using synthetic data generated from a section of the Sigsbee2A model. To further demonstrate the effectiveness of our approach, we test it on an ocean-bottom cable data set from the North Sea. We find that our methodology leads to improved velocity models as evidenced by flatter angle gathers.

Improved 1D velocity model and deep long-period earthquakes in Kanlaon Volcano, Philippines: Implications for its magmatic system

We determined an improved 1D P-wave velocity model for Kanlaon Volcano, one of the most active volcanoes in the Philippines, for use in precise earthquake relocation. The inversion for velocity model was conducted using the program VELEST, which solved the coupled hypocenter–velocity problem from dataset of 251 local earthquakes recorded by the Kanlaon Volcano Seismic Network from 2014 to 2017. Using the improved velocity model, we relocated the volcano-tectonic (VT) earthquakes in Kanlaon with hypoDD, resulting in better-constrained hypocenter locations based on the decrease in RMS and corresponding error assessment tests. Subsequently, we relocated swarms of long-period volcanic earthquakes that occurred from September 1, 2017 to August 31, 2018 and prompted the Philippine Institute of Volcanology and Seismology to raise the volcano's alert level. These long-period earthquakes had emergent arrivals, hence we applied a cross-correlation procedure using REDPy to obtain an improved arrival time picks. These improved hypocenters revealed the existence of deep long-period (DLP) earthquakes that occurred at depths of 23 to 36 km beneath the volcano. These are the first DLP events documented in Kanlaon since the establishment of the Kanlaon Volcano Observatory in 1961. They were found to belong into families based on similarity of waveforms suggesting they were produced by repetitive, non-destructive sources. The occurrences of DLPs in Kanlaon reflected the pressurization of its deep-seated magmatic system. Their identification is critical in assessing the current state of Kanlaon Volcano especially since its last magmatic eruption was in 1902.

Reprocessing legacy three-dimensional seismic data from the Halfmile Lake and Brunswick No. 6 volcanogenic massive sulphide deposits, New Brunswick, Canada

We reprocessed legacy three-dimensional (3D) seismic data from the Halfmile Lake and Brunswick areas, both of which were acquired for mineral exploration in the Bathurst Mining Camp, New Brunswick. Each 3D seismic survey was acquired over known volcanogenic massive sulphide deposits and covered areas with strong mineral potential. Most improvements resulted from a reduction of coherent and random noise on prestack gathers and from an improved velocity model, combined with re-imaging with dip moveout corrections and poststack migration or prestack time migration. At Halfmile Lake, the new imaging results show the Deep zone and a possible extension of the sulphide mineralization at greater depth. True amplitude processing has shown that this anomaly has strong amplitudes and is offset from the Deep zone by a shallowly dipping fault (<15°). With the clearer geological context provided by our results, this anomaly, which appears as a stand-alone anomaly on an original image obtained by Noranda Exploration Ltd., becomes a defendable exploration target. Nonorthogonal acquisition geometry and receiver patches of the Brunswick No. 6 3D seismic survey generated artefacts after dip moveout processing that reduced the overall quality of the seismic volumes. By using a filtering approach based on the application of a weighted Laplacian-Gaussian filter in the Kx–Ky domain, we reduced the noise and improved the continuity of reflections. We also imaged the short and flat reflections observed previously only in the shallow part of prestack time migrated data. These short reflections appear as diffractions on the filtered stacked section with dip moveout corrections, indicating that they originate from small geological bodies or discontinuities in the subsurface.

Application of full waveform inversion algorithms to seismic data lacking low-frequency information from a simple starting model

Full waveform inversion (FWI) is a method that is used to reconstruct velocity models of the subsurface. However, this approach suffers from the local minimum problem during optimisation procedures. The local minimum problem is caused by several issues (e.g. lack of low-frequency information and an inaccurate starting model), which can create obstacles to the practical application of FWI with real field data. We applied a 4-phase FWI in a sequential manner to obtain the correct velocity model when a dataset lacks low-frequency information and the starting velocity model is inaccurate. The first phase is Laplace-domain FWI, which inverts the large-scale velocity model. The second phase is Laplace-Fourier-domain FWI, which generates a large- to mid-scale velocity model. The third phase is a frequency-domain FWI that uses a logarithmic wavefield; the inverted velocity becomes more accurate during this step. The fourth phase is a conventional frequency-domain FWI, which generates an improved velocity model with correct values. The detailed methods of applying each FWI phase are explained, and the proposed method is validated via numerical tests with a SEG/EAGE salt synthetic dataset and Gulf of Mexico field dataset. The numerical tests show that the 4-phase FWI inverts the velocity correctly despite the lack of low-frequency information and an inaccurate starting velocity model both in synthetic data and field data.

Seismicity During the Initial Stages of the Guy‐Greenbrier, Arkansas, Earthquake Sequence

AbstractWe analyze the background seismicity, initiation, and earliest stages of the Guy‐Greenbrier, Arkansas, earthquake sequence, which was potentially induced by wastewater injection starting in July 2010, during the 3 month time period 1 June to 1 September 2010. High‐resolution observations of low‐magnitude seismicity, and the high‐quality Arkansas public well database, facilitate detailed analysis of spatial and temporal correlations between earthquakes, wastewater injection, and hydraulic fracturing. We detected 14,604 earthquakes, with magnitudes −1.5≤ML≤2.9, using two sensitive, waveform similarity‐based event detection methods in parallel: Fingerprint And Similarity Thresholding, and template matching. We located the 1,740 largest earthquakes that form 16 spatially compact clusters, using P and S phases from 3 stations with the double‐difference relocation algorithm and an improved velocity model constrained by the location of quarry blasts. We enhanced the temporal resolution of these event clusters by assigning smaller unlocated events to a cluster based on waveform similarity. Most clustered earthquakes during this time were both spatially and temporally correlated with hydraulic fracturing stimulation at several production wells. For one cluster, microseismicity was correlated with individual stages of stimulation. Many other wells had no detectable nearby seismicity during stimulation. We found a smaller number of events located on the Guy‐Greenbrier Fault that were likely induced by wastewater injection. The concurrent presence of seismicity induced by hydraulic fracturing and wastewater injection presents a challenge for attribution and seismic hazard characterization, but the combination of precision seismology and high‐quality well information allows us to disentangle the effects of these two processes.

Waveform inversion for microseismic velocity analysis and event location in VTI media

Waveform inversion (WI), which has been extensively used in reflection seismology, could provide improved velocity models and event locations for microseismic surveys. Here, we develop an elastic WI algorithm for anisotropic media designed to estimate the 2D velocity field along with the source parameters (location, origin time, and moment tensor) from microseismic data. The gradient of the objective function is obtained with the adjoint-state method, which requires just two modeling simulations at each iteration. In the current implementation the source coordinates and velocity parameters are estimated sequentially at each stage of the inversion to minimize trade-offs and improve the convergence. Synthetic examples illustrate the accuracy of the inversion for layered VTI (transversely isotropic with a vertical symmetry axis) media, as well as the sensitivity of the velocity-analysis results to noise, the length of the receiver array, errors in the initial model, and variability in the moment tensor of the recorded events.

The influence of volcanic rocks on the characterization of Rosebank Field – new insights from ocean-bottom seismic data and geological analogues integrated through interpretation and modelling

Abstract During Late Paleocene–Early Eocene times, the modern Rosebank structure was located at the juxtaposition of the easterly advancing Flett volcanic system and the northerly prograding Flett delta. As a result, the Rosebank reservoir sandstones are interstratified with volcanic and volcaniclastic rocks, offering challenges for reservoir imaging, depth prediction and reservoir characterization. These challenges have driven the application of Ocean Bottom Node (OBN) seismic technology. OBN data have yielded improved velocity models for depth conversion, better reservoir definition and key insights to aid the modelling of sand distribution from seismic attributes. Spectral decomposition of the OBN seismic data has facilitated the extraction of distinct volcanic subunits, whilst spectral enhancement has enabled visualization of complex stacking patterns within individual igneous layers. To complement the seismic analysis, detailed geological analogue studies have been undertaken in volcanic provinces such as the Palaeogene volcanic district of SE Greenland and the Columbia River Flood Basalt Province, USA. No single outcrop provides a definitive analogy to Rosebank, but each offers insights that provide an important link to understanding and managing the main subsurface uncertainties associated with field development. Integration of these multiple workflows have improved the reservoir characterization and provided the foundation for the optimization of the field development plan.

Dual-well 3D vertical seismic profile enabled by distributed acoustic sensing in deepwater Gulf of Mexico

We have acquired and processed 3D vertical seismic profile (VSP) data recorded simultaneously in two wells using distributed acoustic sensing (DAS) during the acquisition of the 2012 Mars 4D ocean-bottom seismic survey in the deepwater Gulf of Mexico. The objectives of the project were to assess the quality of DAS data recorded in fiber-optic cables from the surface to the total depth, to demonstrate the efficacy of the DAS VSP technology in a deepwater environment, to derisk the use of the technology for future water injection or production monitoring without intervention, and to exploit the velocity information that 3D VSP data provide for evaluating and updating the velocity model. We evaluated the advantages of DAS VSP to reduce costs and intrusiveness, and we determined that high-quality images can be obtained from relatively noisy raw 3D DAS VSP data, as evidenced by the well 1 image, probably the best 3D VSP image we have ever seen. Our results also revealed that the direct arrival traveltimes can be used to assess the quality of an existing velocity model and to invert for an improved velocity model. We identified issues with the DAS acquisition and the processing steps to mitigate them and to handle problems specific to DAS VSP data. We described the steps for conditioning the data before migration, reverse time migration, and postmigration processing to reduce noise artifacts. We outlined a novel first-break picking procedure that works even in the absence of a strong first arrival and a velocity diagnosis method to assess and validate velocity models and velocity updates. Finally, we determined potential applications to 4D monitoring of fluid movement around producer or injector wells, identification of active salt movements, and more accurate imaging and monitoring of complex structures around the wells.

Using reverse vertical seismic profiling (RVSP) to characterise the subsurface fracture system of the Seokmo Island geothermal field, Republic of Korea

This paper presents a case study of fracture-system interpretation using reverse vertical seismic profiling (RVSP) with seismic reflection, refraction, and borehole televiewer data on Seokmo Island, South Korea. We first extracted fracture locations from a pre-existing image obtained by prestack phase-screen migration of RVSP data, and the strike and dip of each fracture from the borehole televiewer data. We established an initial velocity model using this fracture information and then generated synthetic common-receiver gather data through forward simulation. However, the synthetic data could not sufficiently reflect the characteristics of the field data. To resolve this problem, we added an upper alluvial layer, observed in the surface reflection and refraction data, to the velocity model. To improve the quality of the migrated image of the RVSP field data and the velocity model, we not only reprocessed the RVSP data, but also applied prestack generalized-screen migration, which is more accurate than phase-screen migration in computing steep fractures. The new synthetic data generated from the improved velocity model agreed well with the RVSP field data. As a result, we could describe a markedly improved subsurface structure including fracture locations. The workflow suggested in this study will be helpful for imaging fracture systems in oil and gas reservoirs as well as in geothermal reservoirs.

Waveform tomography of field vibroseis data using an approximate 2D geometry leads to improved velocity models

Waveform tomography, a combination of traveltime tomography (or inversion) and waveform inversion, is applied to vibroseis first-arrival data to generate an interpretable model of P-wave velocity for a site in the Nechako Basin, south-central British Columbia, Canada. We use constrained 3D traveltime inversion followed by 2D full-waveform inversion to process long-offset (14.4 km) first-arrival refraction waveforms, resulting in a velocity model of significantly higher detail than a conventional refraction-statics model generated for a processing workflow. The crooked-line acquisition of the data set makes 2D full-waveform inversion difficult. Thus, a procedure that improves the tractability of waveform tomography processing of vibroseis data recorded on crooked roads is developed to generate a near-surface ([Formula: see text]) velocity model for the study area. The data waveforms are first static corrected using a time shift determined by 3D raytracing, which accounts for the crossline offsets produced by the crooked-line acquisition. The velocity model generated from waveform tomography exhibits substantial improvement when compared with a conventional refraction-statics model. It also shows improved resolution of sharp discontinuities and low-velocity regions when compared to the model from traveltime tomography alone, especially in regions where the geometry errors are moderate. Interpretation of the near-surface velocity model indicates possible subbasins in the Nechako Basin and delineates the Eocene volcanic rocks of the study area. This approach limits the ability of the full-waveform inversion to fit some propagation modes; however, the tractability of the inversion in the near-surface region is improved. This new development is especially useful in studies that do not warrant 3D seismic acquisition and processing.

P, S wave velocity model of the crust and upper most mantle of Albania region

This paper describes the one-dimensional (1D) velocity model computed by VELEST in the SEISAN seismic analysis system, inverting re-picked P-wave and S-wave arrival times recorded during 2002–2006 by the Albanian, Montenegro, Thessalonica and Macedonia seismic networks. The re-picked data yield P-wave and S-wave velocities proved to be more suitable compared to bulletin data for this detailed inversion study. Seismic phases recorded by the Albania seismic network and integrated with data from the Montenegro, Thessalonica and Macedonia networks are used to prepare the Albanian seismic bulletin. Earthquake hypocenters from the Albanian bulletins have also location errors that are negligible for civil protection purposes, large scale seismotectonic analyses and more accurate hypocentral determinations which are necessary for detailed seismotectonic and geodynamic studies. It was noted that the smoothness of the velocity variation increased with depth. A velocity of 5.5 km/s was calculated for the upper crust, 6.1 km/s was calculated for the middle crust and 6.9 km/s was computed for the lower crust. P wave velocity was 7.85 km/s at depth of 50 km and for the upper mantle it is 8.28 km/s. Using the improved velocity model, the earthquakes which occurred in Albania in the past 5 years were able to be relocated, achieving constrained hypocentral determinations for events in Albania. The interpretation of the 1 D velocity models infers interesting features of the deep structure of Albania. These results represent an important step towards more detailed seismotectonic analyses.

Seismicity in the Rotorua and Kawerau geothermal systems, Taupo Volcanic Zone, New Zealand, based on improved velocity models and cross-correlation measurements

We analyze earthquakes occurring in and around the Rotorua and Kawerau geothermal systems, Taupo Volcanic Zone, New Zealand. The two data sets contain 504 and 1875 shallow (≤ 20 km deep) earthquakes, respectively, and span the 21 year period between 1984 and 2004. The arrival time data for these earthquakes are first used to calculate 1-D P- and S-wave seismic velocity models and accompanying station correction terms for both areas. In order to address the non-uniqueness of the joint hypocenter-velocity model estimation problem, we analyze suites of 1000 velocity models computed from random initial models. The final velocity models are well constrained, particularly at depths between 4 and 15 km, and consistent with the results obtained in previous seismic refraction studies of the central Taupo Volcanic Zone. Using a combination of cross-correlation-derived and catalog-based arrival times, we relocate subsets of the Rotorua and Kawerau data sets. In Rotorua, the relocated earthquakes cluster near the geothermally active parts of Rotorua City and beneath the Mount Ngongotaha rhyolite dome. Earthquake clusters and alignments reveal seismogenic structures in the mid-crust whose positions and geometries are consistent with previously published fault mechanisms and known near-surface faults. In Kawerau, the earthquakes within the geothermal field align along northeast-trending lineations, consistent with the predominant alignment of surface-mapped faults in the area.

Salt interpretation enabled by reverse-time migration

Many production targets in greenfield exploration are found in salt provinces, which have highly complex structures as a result of salt formation over geologic time. Difficult geologic settings, steep dips, and other wave-propagation effects make reverse-time migration (RTM) the migration method of choice, rather than Kirchhoff migration or other (by definition approximate) one-way equation methods. Imaging of the subsurface using any depth-migration algorithm can be done successfully only when the quality of the prior velocity model is sufficient. The (velocity) model-building loop is an iterative procedure for improving the velocity model. This is done by obtaining certain measurements (residual moveout) on image gathers generated during the migration procedure; those measurements then are input into tomographic updating. Commonly RTM is applied around salt bodies, where building the velocity model fails essentially because tomography is ray-trace based. Our idea is to apply RTM directly inside the model-building loop but to do so without using the image gathers. Although the process is costly, we migrate the full frequency content of the data to create a high-quality stack. This enhances the interpretation of top and bottom salt significantly and enables us to include the resulting salt geometry in the velocity model properly. We demonstrate our idea on a 2D West Africa seismic line. After several model-building iterations, the result is a dramatically improved velocity model. With such a good model as input, the final RTM confirms the geometry of the salt bodies and basically the salt interpretation, and yields a compelling image of the subsurface.

Oceanic crust and Moho of the Pacific Plate in the eastern Ogasawara Plateau region

Abstract To show the structure of oceanic crust and Moho around the eastern Ogasawara Plateau, we have analyzed industry‐standard two‐dimensional multichannel seismic reflection data. To obtain improved velocity models, phase information of seismic signals was used for velocity analysis and velocity models for oceanic crust above Moho were determined. We apply this velocity analysis technique to seismic reflection data around the eastern Ogasawara Plateau, with the result of clear images of structures within oceanic crust and Moho. South of the Ogasawara Plateau, Moho deepens proximal to the Plateau. Moho distal to the Plateau is ca 7 km below sea floor (bsf), whereas it is ca 10 km bsf near the Plateau. The characters of oceanic crust and Moho differ significantly north and south of the Plateau. To the north, the structure of oceanic crust is ambiguous, the sea floor is shallower and less smooth, and Moho is discontinuous. To the south, structures within oceanic crust and Moho are imaged clearly, and the sea floor is deeper. A strong Moho reflection south of the Plateau might represent a sharp boundary between layered gabbro and peridotite. However, discontinuous Moho reflections north of the Plateau might represent rough topography because of intensive magmatism or a gradual downward increase in velocity within a thick Moho transition zone. A fracture zone north of the Plateau also appears to separate oceanic crust and Moho of different characters, suggesting vigorous magmatism between the Plateau and the fracture zone, and that the Ogasawara Plateau and the fracture zone influenced the genesis of oceanic crust and upper mantle. Differences in acoustic characteristics to the north and south of the Plateau are apparent in profiles illuminated by seismic attributes.

Modeling and inversion of PS-wave moveout asymmetry for tilted TI media: Part 2 — Dipping TTI layer

Dipping transversely isotropic layers with a tilted symmetry axis (TTI media) cause serious imaging problems in fold-and-thrust belts and near salt domes. Here, we apply the modified [Formula: see text] method introduced in Part 1 to the inversion of long-offset PP and PS reflection data for the parameters of a TTI layer with the symmetry axis orthogonal to the bedding. The inversion algorithm combines the time- and offset-asymmetry attributes of the PSV-wave with the hyperbolic PP- and SS-wave moveout in the symmetry-axis plane (i.e., the vertical plane that contains the symmetry axis). The weak-anisotropy approximations for the moveout-asymmetry attributes, verified by numerical analysis, indicate that small-offset (leading) terms do not contain independent information for the inversion. Therefore, the parameter-estimation algorithm relies on PS data recorded at large offsets (the offset-to-depth ratio has to reach at least two), which makes the results generally less stable than those for a horizontal TTI layer in Part1. The least-resolved parameter is Thomsen’s coefficient [Formula: see text]that does not directly influence the moveout of either pure or converted modes. Still, the contribution of the PS-wave asymmetry attributes helps to constrain the TTI model for large tilts [Formula: see text] of the symmetry axis [Formula: see text]. The accuracy of the inversion for large tilts can be improved further by using wide-azimuth PP and PS reflections. With high-quality PS data, the inversion remains feasible for moderate tilts [Formula: see text], but it breaks down for models with smaller values of [Formula: see text] in which the moveout asymmetry is too weak. The tilt itself and several combinations of the medium parameters (e.g., the ratio of the P- and S-wave velocities in the symmetry direction), however, are well constrained for all symmetry-axis orientations. The results of Parts 1 and 2 show that 2D measurements of the PS-wave asymmetry attributes can be used effectively in anisotropic velocity analysis for TTI media. In addition to providing an improved velocity model for imaging beneath TTI beds, our algorithms yield information for lithology discrimination and structural interpretation.

Three‐dimensional structure from feathered two‐dimensional marine seismic reflection data: The eastern Nankai Trough

For a typical two‐dimensional (2‐D) marine multichannel seismic reflection survey, cross currents cause receiver cable side drift, and data collection becomes a limited swath 3‐D survey. Processing of swath marine survey data using standard 2‐D imaging procedures can create spurious discontinuities and wipeouts in the images of reflection events. These anomalies are commonly attributed to geology and tectonics during interpretation. We show how discontinuous reflections and wipeouts are created wherever subsurface structures have significant cross‐profile dip. We also show that more accurate imaging can be achieved by cross‐dip processing, which takes into account the 3‐D character of the data and of the geology. The final products are a much improved stack, determinations of structural dips at right angles to the survey line, and an improved velocity model. We apply this cross‐dip processing to data from the eastern Nankai Trough, SW Japan. Improved imaging combined with the new cross‐dip information allows us to constrain better the 3‐D character of the deformation in the accretionary prism and oceanic crust. In particular, the extracted structural cross‐dip values provide evidence that supports two hypotheses related to the underthrusting of a large oceanic ridge (Paleo‐Zenisu ridge) underneath the accretionary wedge: (1) Parts of the accretionary wedge basins have been uplifted and deformed due to the subduction of a volcano on the ridge. (2) The early phase of oceanic lithosphere shortening, which produced the underthrusted Paleo‐Zenisu ridge, was related to N‐S compression oblique to the general plate motion in this region.

Double-Difference Tomography: The Method and Its Application to the Hayward Fault, California

We have developed a double-difference (DD) seismic tomography method that makes use of both absolute and relative arrival times. By reducing systematic errors using the more accurate relative arrival times, the method produces an improved velocity model. Simultaneously, it yields event locations of a quality equivalent to those of the DD location method, while avoiding simplifying assumptions of that method. We test this method on a synthetic dataset and find that it produces a more accurate velocity model and event locations than standard tomography. We also test this method on a Hayward fault, California, earthquake dataset spanning 1984–1998. The earthquakes relocated by this method collapse to a thin line along the fault trace, consistent with previous results. The DD velocity model has sharper velocity contrasts near the source region than the standard tomography model.