Gordon Poole

RTM Noise Attenuation and Image Enhancement Using Time-shift Gathers

Reverse time migration (RTM) has become a central tool in seismic imaging, especially in complex geologies where other migration methods fail. However, the way RTM is generally formulated introduces various types of artifacts, these include low-frequency noise in high velocity-contrast areas and shear-wave-like noise when anisotropy is included in the formulation. Using time-shift gathers from the extended imaging condition, we develop a method to effectively attenuate these artifacts and further enhance the final image by refocusing the migrated energy. We demonstrate the results of our method on both real and synthetic data sets.

Deblending of simulated simultaneous sources using an iterative approach: an experiment with variable-depth streamer data

We have simulated a simultaneous source experiment with a real variable-depth streamer NAZ survey acquired in Brazil and applied an iterative separation flow using a hybrid of median and f-x projection filters for deblending. The results show little leakage after deblending in both near and far offsets while preserving the AVO characteristics. We have also performed Kirchhoff migration in order to examine the residual cross-talk in common image gathers after deblending. The results reveal little leakage, and the quality of the common image gathers is adequate for velocity model building. We also study how the existence of spatial aliasing when shots are coarsely sampled degrades the deblending quality. Lastly, the impact of firing multiple sources simultaneously to improve shot sampling is investigated.

Simultaneous multivintage time-shift estimation

Time-shift estimation is a key step in seismic time-lapse processing as well as in many other signal-processing applications. We consider the time-shift problem in the setting of multiple repeat surveys that must be aligned consistently. We introduce an optimized least-squares method based on the Taylor expansion for estimating two-vintage time shifts and compare it to crosscorrelation. The superiority of the proposed algorithm is demonstrated with synthetic data and residual time-lapse matching on a U. K. continental shelf data set. We then discuss the shortcomings of cascaded time alignment in multiple repeat monitor surveys and propose an approach to estimate simultaneous multivintage time shifts that uses a constrained least-squares technique combined with elements of network theory. The resulting time shifts are consistent across all vintages in a least-squares sense, improving overall alignment when compared to the classical flow of alignment in a cascaded manner. The method surpasses the cascaded approach, as noted with sample synthetic and three-vintage U. K. continental shelf time-lapse data sets.

Processing Solutions For Wide Azimuth Data: Outcome From a WATS Field Experiment In Deep Water Gulf of Mexico

Wide Azimuth Towed Streamer (WATS) acquisition has already proved to be a key technique in improving seismic imaging, especially in complex areas such as sub-salt plays. In 2006, a WATS field experiment was conducted in a deep water area of the Gulf of Mexico. The main purpose was to challenge recently developed 3D processing algorithms and find the most suitable processing strategy for a wide azimuth dataset. The results indicate that a 3D shot based processing sequence is an effective solution that accommodates the effects related to the multi-pass acquisition method and realizes the full benefit of the recorded 3D wide azimuth wave field.

Shot-based pre-processing solutions for wide azimuth towed streamer datasets

Wide azimuth towed streamer (WATS) acquisition has been shown to provide improved seismic imaging, especially in areas with complex 3D structures. By making use of additional source vessels shooting into the streamer array from large lateral offsets, a dataset is created with a large cross-line aperture, higher fold, and a broader offset-azimuth distribution than conventional (narrow azimuth) streamer datasets. The improved imaging provided by WATS acquisition geometry has been well illustrated as this method is being more widely adopted. An example which illustrates the superiority of WATS data for imaging was given by Michell et al. (2006). There, a straightforward depth migration of WATS data with limited pre-processing was shown to provide a significantly improved image compared to the results obtained with conventional streamer acquisition. However, to realize the full potential of WATS data, preprocessing (the processing sequence prior to imaging) is essential, especially for applications which depend on the quality of the pre-stack gathers such as: N Velocity model building and update N Pre-stack time and depth imaging N 4D/time lapse processing N AVO analysis and reservoir characterization N Quantitative analysis

Simultaneous Multi-Vintage Multi-Parameter Time-lapse Matching

D matching is an important processing step which is repeatedly used within a time-lapse processing sequence in order to remove residual amplitude, time and phase differences between vintages of seismic data. In this paper we introduce a new matching algorithm that minimizes the global NRMS between an arbitrary number of vintages and thus automatically achieves the best possible overall repeatability in one single matching step. The algorithm can work in the time- or frequency domain and simultaneously finds all matching parameters for all vintages. By using constraints a unique solution to this non-linear optimisation problem is found without the necessity to define an upfront reference vintage. This minimises the risk of propagating artefacts which may be present on one of the vintages to all other datasets. We show two example applications of the new algorithm. Firstly, we use the new algorithm for pre-imaging sail-line consistent removal of acquisition related artefacts (destriping) of multiple vintages. Secondly, we present examples of multi-vintage matching, as applicable to 4D residual local matching on imaged data.

A Regularization Workflow for the Processing of Cross-spread COV Data

Data regularization is critical for the suppression of Kirchhoff migration noise and the production of a clean migration image. We introduce a workflow utilizing Fourier reconstruction in two spatial dimensions to fully regularize cross-spread COV data i

Multi-Dimensional Data Regularization for Modern Acquisition Geometries

P143 Multi-Dimensional Data Regularization for Modern Acquisition Geometries G. Poole* (CGGVeritas) & P. Herrmann (CGGVeritas) SUMMARY Data regularization is critical for the suppression of Kirchhoff migration noise and the production of a clean migration image. Techniques that perform data regularization simultaneously along two axes provide a solution where multiple passes of a 1D approach fail. We introduce a versatile two-dimensional Fourier reconstruction algorithm that regularizes the input data as well as filling gaps in the coverage. We validate the algorithm on a synthetic cross-spread gather example as well as demonstrating the technology on a real offset volume dataset. The results

Shot Based Pre-Processing Solutions for a WATS Survey – An Example from a Field Trial in Green Canyon – Gulf of Mexico

C014 Shot Based Pre-Processing Solutions for a WATS Survey – An Example from a Field Trial in Green Canyon – Gulf of Mexico M. Magesan (CGGVeritas) J.-C. Ferran* (CGGVeritas) S. Kaculini (CGGVeritas) C.J. Faulkner (CGGVeritas) P. Herrmann (CGGVeritas) A. Pica (CGGVeritas) G. Poole (CGGVeritas) & S. Le Roy (CGGVeritas) SUMMARY In July 2006 a 3D Wide Azimuth Towed Streamer (WATS) field trial was carried out in a deep water area of Green Canyon in the Gulf of Mexico. The purpose was to provide insight into the potential problems and complexities of such seismic datasets and to challenge conventional and newly