Largeo

English

Innovative Technologies

  • PDF

CRS Processing

TEECsolutions and LARGEO Form Strategic Partnership in Russia

The Germany based company TEECsolutions and Russian company LARGEO have announced a strategic partnership on high quality seismic imaging in Russia. TEECsolutions is the international branch of TEEC, the leading company for CRS (Common Reflection Surface) seismic processing. The CRS technique is providing intelligent data interpolation techniques honoring local dip and azimuth, and is therefore superior to all techniques relying on borrowing or duplicating traces only. Furthermore, the signal-to-noise ratio is significantly increased during this processing step leading to superior imaging quality of subsequent prestack migration.

From this alliance it is expected that the use of TEEC' s proprietary CRS techniques enables LARGEO to improve the structural imaging in highly complex geological settings, and to offer more accurate reservoir geophysical studies.

Additional resources:

Multiples Attenuation Technology

Surface Related Multiple Elimination Method in Offshore Seismic Data Processing

Multiples is a major challenge in offshore seismic data processing. Multiple attenuation results have key impact on further imaging and interpretation of offshore seismic data/ Building of correct velocity models, which are extremely important in seismic imaging, is strongly depends on multiple elimination quality. Modern tomography approaches for velocity modeling do not separate true and multiple reflections. The quality of the migration is directly related to the quality of the velocity model. It also affects structural interpretation and reservoir analysis.

The most popular multiple-predictive technique is Surface Related Multiple Elimination method (SRME). The SRME does not require geological parameters or velocity characteristics. This will ensure minimizing processing errors, lower computer time requirements and higher reliability and accuracy of the results. The sequence of generating noise models is based on the mathematical operation of trace autoconvolution.  Multiple iterations offer the most accurate and robust solution, thus, avoiding removal of primary reflactions. Therefore, all that is required for efficient, high quality and safe multiple attenuation is to ensure compliance with the applicable technical standards and procedures, and optimization of computing power.

Field of seismograms with multiples: a) Seismogram before attenuation; b) the multiple model obtained using 3D SRME; c)The result of subtracting a multiple of the wave.

The modeling can be done in 2D or 3D versions. Modeling is performed in 2D and 3D. 2D modeling approach is used for simple cases such as simple seabed topography and reflecting boundary configurations. It is also applied when multiples are shifted along the seismic line (seabed topography and reflector dipping in the direction of the sail line). The 2D models are more efficient in terms of time required to build the model. It shall be noted that SRME is very efficient in predicting all free surface multiple energy in the near offset stacks that other conventional methods, based on separation of fields using reflection event curvature, cannot handle. The most accurate and robust multiple model can be obtained using a 3D modeling approach.

Single-channel cross section a) before suppressing of multiplesb) after the suppression of multiples  using 2D SRME and c) after the suppression of multiples using 3D SRME.

For 2D surveys we apply multiiterational 2D SRME approach. The algorithm das be described as follows: initially, modeling and adaptive subtraction of the model is conducted. The results are used for further modeling to build the second iteration of the multiple model. The second iteration is added back to the first iteration. The resulting model is subjected to further modeling. The multi-iteration approach allows to build an integrated model for further adaptive subtraction from the original seismic data. The Multi-Iteration Technique is more efficient in attenuating multiple energy in complex environments andsuppression of the multiples of higher order.

Input data (a), 1st iteration 2D SRME (b), 2nd iteration 2Д SRME (v).

5D Regularization of Marine Data

Analysis of Seismic Data Regularization Results Applied to Real Marine Data (Black Sea and Okhotsk Sea Data)

Uniform distribution of traces over the survey area and of the offsets within each bin is very important for offshore data processing. During marine operations seismic acquisition tests typically control an average fold and at best – far, near and middle offset distribution within bins. Acquired seismic datasets are irregularly sampled in offset.

Due to rough sea condition and cable feathering effect marine seismic data processing mostly deal with irregularly acquired surveys. Industry Kirchhoff migration algorithms assume the input data consist of one or more offset planes, each made up of traces with constant offset and azimuth, falling on a perfectly regular grid of mid-point locations . Variable foldage with clusters of closely spaced traces and gaps in coverage has adverse effect on velocity estimation, noise and multiples suppression and can produce severe migration artifacts . There are several processing techniques to minimize the effect of irregularly sampled data, and migration results can strongly depend on the type of the algorithm applied.

Our company generally applies two main approaches to condition seismic data prior to migration.

  1. Fold compensation is done through the geometrical weighting calculated as reciprocal of number of traces in the bin. The operation allows weighing of selective offset and azimuth ranges and can be advantages minimizing migration artifacts, increasing the accuracy of migration.
  2. The regularization is the processing technology that attempts to adjust the recorded data to more nearly match the assumptions of the migration algorithms, with a consequent improvement in amplitude fidelity and reduction in noise.

The regularization workflow can be described as follows:

  • Sort data to: common offset plane / cross line / inline
  • Within each offset:
    • Apply partial NMO to nominal offset
    • Optionally reject redundant traces (minimum azimuth deviation / distance from bin centre)
    • Apply regularization to each cross line
    • Optionally apply noise removal such as FXYdecon
    • Remove partial NMO
  • Sort back to 3D CMP gathers for imaging

Comparison of cross-line section. А) PreSDM results without fold compensation procedures, B) PreSDM Results with Fold Compensation, C) PreSDM results after Data Regularization

The regularization of marine data is resource- consuming operation because of massive amount of high-fold data, it requires disc space four times more then was occupied by input data.

A) PreSDM Results with Fold Compensation

B) PreSDM results after Data Regularization

Application of Anisotropic Tomography

Application of Anisotropic Tomography Solution for Generation of Accurate Subsurface Images.

Isotropic migration has significant limitations that prevent effective imaging in some instances, since it has been performed with the assumption of hyperbolic normal move-out (NMO) for horizontal reflector travel time in seismic data processing. Anisotropy is the variation of seismic waves velocities as a function of traveling direction. For multilayered subsurface (sands and shales) the acoustic waves velocities within the layers is higher then the velocities in the orthogonal direction. The anisotropic migration overcames the limitations of the conventional migration. For anisotropy correction we introduce two anisotropic parameters “epsilon” and “delta”. The Delta parameter is calculated from mis-ties of depths between well-data and migrated seismic data. The Epsilon parameter is the result of residual normal moveout analysis at the far offsets, or tomographic inversion; when well data are not available, the anisotropy correction can be made using only “epsilon” parameter. Employment of anisotropy parameters gives us a significant improvement in final image quality, especially in  complex geological environments with mixed-lithology fault zones, gas chimneys presence and subtle facies variations.

 

The anisotropic solution provides more accurate interval velocities analysis and increases AVO capabilities due to utilization of voluble information in seismic data. LARGEO applies GXT iterative hybrid tomography method for building of anisotropic velocity model.

Comparison А) – isotropic and B) – anisotropic interval velocities models

 

Duplex Wave Migration

Conventional CDP processing does not give unambiguous solution for imaging of salt wall zero or near zero throw faults within the reservoir (fault compartmentalization), zones of fracturing. The main reason for this is a limited capability of Conventional pre-stack depth migration (PSDM) for imaging of vertical boundaries. The first reflections of waves from this boundaries, do not reach the registration surface. The DWM algorithm is designed to image the DWE that will arrive at a time greater than that of the primary base boundary. A beam tube construction eliminates the migration noise that would result from including the base boundary primary reflections in the migration summation. Tight control of the aperture is also key to suppression of artifacts from primary reflections. Each DWM run produces four separate and distinct views of the vertical boundaries based on two possible bounce orders – base boundary then vertical boundary or vice versa and traces input to the migration are either to the right of the shot or to the left of the shot.

Duplex Wave Migration is based on the Kirchhoff transformation, where the Green’s function is changed according to the travel time characteristics (kinetics) of the duplex wave (double reflection: from the sub-horizontal and sub-vertical boundaries and vice versa).

A primary event is defined in depth by the user and must be deeper than the vertical boundaries we wish to image. This boundary should be one of the layers described in the 3D macro model of the medium. It is assumed that a conventional Pre-stack Depth Migration (PSDM) has been run prior to running Duplex Wave Migration (DWM), therefore we use the depth model generated from that process as a starting point. The depth model can be either isotropic or TTI anisotropic. Traces input to the migration are either to the right of the shot or to the left of the shot. Since we have an image of the same vertical boundary from both sides of the boundary, they will only focus at the same location in 3D space if the velocities are correct. In order to trace the sub-vertical boundaries at the level of the target horizons, usually, several base layers are provided and the resulting images are interpreted jointly.

The company has already performed several 2D and 3D seismic projects with Timano-Pechora and west Siberia datasets with the objective to asses and delineate fracturing zones in Bazhenov and Jurassic formations.

Example of DWM results revealing amplitude anomalies in Bazhen formations. The anomalous values indicate perspective fracturing zones within Bazhen formation

Comparison of conventional migration results with DWM for mapping of tectonic destructions.

You are here: Technologies