Date

Group

Speakers

Summary

9-6

S-Team

Clement and Roel

9-13

I - Team

Francesco

Computation of the image perturbation for WEMVA in the shot-domain

Migration velocity analysis is based on the semblance principle: different experiments 'see' the same earth. In other words, the images we obtain from different shots must be consistent with each other, if the velocity model we use for migration is correct. This consistency is usually evaluated once that all the experiments have been imaged; this is expensive and becomes even more costly if the imaging procedure we use is based on a time-domain finite difference scheme. Moving from a measure of consistency between migrated images, in this presentation I address the problem of computing image perturbations for driving the WEMVA step and recovery the velocity error information.

9-20

C - Team

Simon

Waveform Inversion, Wavefield Decomposition, and Wavenumber Filtering

9-27

A - Team

Ibrahim Mustafayev (RCP) and Mamoru

Ibrahim: Lithology discrimination in Delhi Field, Louisiana

The talk will be focused on the successful application of (AVA) inversion of prestack - seismic amplitude data to detect and delineate sandstone reservoirs in the Delhi Field.

Mamoru: Anisotropic MVA in the presence of a quadratic velocity variation

To efficiently handle lateral heterogeneity with the scale smaller than spreadlength, I employ a factorized VTI model with a quadratic velocity variation. I show that the quadratic velocity variation in the shallow layer primarily contributes to estimates of medium parameters for the deep layers.

10-4

S - Team

Filippo

The contribution of the spatial derivatives to surface wave interferometry - A study on real and synthetic data.

We designed a field experiment to assess whether or not the use of spatial derivatives can add value to the estimates made using virtual-receiver interferometry. This is a data driven project, we first analyze the data acquired in the field and then we use modeling to act as a verification of what we observe. This workflow differs from the usual practice, where new applications are tested first on syntethic data and only after on real data.

10-6

Guest Speaker

Prof. Serge Shapiro (Free University of Berlin, Germany)

Quantitative understanding of microseismicity for reservoir characterization and development

Characterization of fluid-transport properties of rocks is one of the most important, yet most challenging goals of reservoir geophysics. There are some fundamental difficulties related to using active seismic methods for estimating fluid mobility and rock permeability. Borehole fluid injections are often applied for the stimulation and development of hydrocarbon and geothermal reservoirs. Recently, it has become clear that understanding and monitoring fluid-induced seismicity is necessary for hydraulic characterization of reservoirs as well as for robust assessment of reservoir stimulations and related seismic risk.This lecture gives an introduction into the main principles of seismicity-based reservoir characterization.

10-11

I - Team

Jeff

Improving Blended Source Imaging

New insights into blended source imaging may allow us to substantially improve the image quality of seismic imaging while also reducing the cost of imaging. Past work has been on developing new schemes for imaging, but new work suggests that better results may be possible by extending previously developed schemes (such as plane-wave migration). We'll take a look at how we can improve some of these schemes, and possible future developments for blended source imaging.

10-18

Fall Break/SEG

No Seminar

10-25

SEG Debriefing

11-1

C - Team

Luming

Eliminating artifacts in fault displacement fields

The estimated fault displacement field contains lots of artifacts. This talk shows both the efforts I made in this during the summer and remaining problems.

11-8

A - Team

Steve

VSP-based characterization and inversion of a multi-layered isotropic/VTI reservoir overburden

I present a case study for modeling a Vertical Seismic Profile (VSP) survey of a typical tight-gas reservoir with known fracture directions from previous seismic processing. Typical seismic models for western US oil and gas reservoirs consist of one or more layers of Horizontally Transverse Isotropic (HTI) material with isotropic overburdens. Migration velocity model improvements may be obtained by investigating and inverting for the structure, velocity, and any Vertically Transverse Isotropic (VTI) layers in the overburden. Highly detailed velocity models may be derived from VSP data, and modeling/inversion quality is directly effected by the ability to accurately constrain near-surface properties. We employ ANRAY ray tracing software to develop example ground truth data sets for testing a ray-based, tomographic, sparse-least squares (LSQR) inversion algorithm. The process is used to invert for P-wave velocity and anisotropy parameters δ and ε. I will discuss the performance of ANRAY in this specific case (and its ramifications to similar/future work), and close with a discussion of the performance of the inversion algorithm with respect to modeled and actual VSP arrival times.

11-15

S - Team

Kanu

Investigation of a dynamic creep response on the Southern Hayward Fault Triggered by the 1989 Loma Prieta, California Earthquake

Theodolite measurements across the right-lateral Hayward fault, San Francisco Bay, California show a dramatic reduction in surface creep rate from 5-10 mm/yr before the 1989 Loma Prieta earthquake to nearly zero creep rate after the earthquake. This reduction in surface creep which lasted for about 6-year was followed by a fast sudden creep event that accumulated about 20-25 mm of right-lateral displacement within the Southern section of the fault. In this study, we show that this fault creep behavior can be explained by a sudden reduction in shear stress on a fault imbedded in an elastic medium whose slip is governed by laboratory-derived friction laws. We infer slip-governing parameters the describes the fault properties using this creep events. These parameters include, frictional rheology parameters, depth extent of surface creep, effective normal stress, etc.

11-22

I -Team

Tom and Pengfei

Tom: An image-guided method for automatically picking CIP locations

By using CIPs instead of CIGs, we are free to analyze migrated images at locations that are as sparse as necessary and that are aligned with geological structure. This approach reduces computation cost. Unfortunately, automating the construction of CIPs is a tougher task than it is to automate the construction of CIGs. To take advantage of the computational cost savings that CIPs can offer, they need to be constructed along image locations that correspond to geological features. I describe a semi-automatic method for picking CIP locations from seismic images. Our method uses local image properties including structure-oriented semblance, linearity, and the analytical envelope to compute a priority map of seismic images that indicates the best locations to construct CIPs. Once priority map is computed, we pick CIP locations that are distributed sparsely throughout the image and along reflectors by greedy heuristic.

Pengfei: A refined harmonic Arnoldi method for computing eigenvalues of large matrices.

The harmonic Arnoldi method can be used to compute some eigenvalues of large matrices.But the eigenvectors obtained by the method may converge erratically and even fail to converge.This talk will describe a refined harmonic Arnoldi method that improves convergence and is more attractive.

11-29

C - Team

Chris and Farhad

Chris: A closer look at 3D painting

3D painting is designed in a way to be intuitive and easy to run. Let's discuss how to construct a 3D painting and demonstrate techniques I used to paint a 3D salt body.

Farhad: Using image-guided Interpolation in determining the Geometry of a Subducting Plate

More precise knowledge of the geometry of subducting slabs could affect the result of Earthquake source inversion and hazard analysis based on such results. I study the possibility of using the Image-guided interpolation method in constraining the geometry of subducting plates.

12-6

A - Team

Bharath

Attenuation analysis for a cross-hole, perforation data set

Recently we have received from Shell E&P, crosshole data suitable for robust estimation of P-wave attenuation coefficients. The wavefield was generated in a tight-gas reservoir by perforation shots fired at regular intervals in a horizontal well. The data were recorded by 11 multicomponent receivers placed at different depths in a vertical (monitor) well. We will present and discuss a methodology to obtain frequency independent P-wave attenuation coefficients and discuss possible anisotropic attenuation signatures.