Intelligent stacking techniques become one of the main keys to obtain migrated images with enhanced signal-to-noise ratio and to improve imaging quality. The classical stacking methods, that equally considers each individual traces, reach their limits when they have to deal with unequally sampled illumination like in any subsalt environments. They are not sufficient to suppress coherent noise nor to highlight weaker signals. The emergence of Wide-Azimuth acquisitions has largely contributed to improve the illumination, the signal-to-noise ratio and helps the velocity model building in complex areas. This large amount of data has however to be carefully treated to bring out events of interest. In this work, we describe a methodology taking advantages of our recent progresses in Reverse Time Migration to select data contributing the most to a given target. Even if this technique could be easily applied in a pre-migration step, the advantage of the proposed method is to use it post-migration in order to avoid steadily re-migrating the input data. Presentation Date: Thursday, October 18, 2018 Start Time: 8:30:00 AM Location: 207A (Anaheim Convention Center) Presentation Type: Oral

Coping with the raise of data volume is becoming critical in almost all industries. In petroleum industry, especially in seismic where the tendency is to acquire denser data and migrate finer image, methods have to be put in place in order to mitigate this trend. In the paper we discuss about the use of lossy data compression in a Gaussian beam migration algorithm to handle smaller input volume and temporary data. To better represent seismic data, wavelet packet transform technics are used as a basis for the compression algorithm. Thanks to this technique we can achieve good compression ratio without affecting the data quality. Finally, the compression algorithm had to be redesigned to be able to run efficiently enough to highly compete to non-compressed data access. Presentation Date: Wednesday, October 17, 2018 Start Time: 1:50:00 PM Location: Poster Station 20 Presentation Type: Poster

Least-squares migration (LSM) can produce images with better balanced amplitudes and fewer artifacts than standard migration. The conventional objective function used for LSM minimizes the L2-norm of the data residual between the predicted and the observed data. However, for field-data applications in which the recorded data are noisy and undersampled, the conventional formulation of LSM fails to provide the desired uplift in the quality of the inverted image. We have developed a least-squares reverse time migration (LSRTM) method using local Radon-based preconditioning to overcome the low signal-to-noise ratio (S/N) problem of noisy or severely undersampled data. A high-resolution local Radon transform of the reflectivity is used, and sparseness constraints are imposed on the inverted reflectivity in the local Radon domain. The sparseness constraint is that the inverted reflectivity is sparse in the Radon domain and each location of the subsurface is represented by a limited number of geologic dips. The forward and the inverse mapping of the reflectivity to the local Radon domain and vice versa is done through 3D Fourier-based discrete Radon transform operators. The weights for the preconditioning are chosen to be varying locally based on the relative amplitudes of the local dips or assigned using quantile measures. Numerical tests on synthetic and field data validate the effectiveness of our approach in producing images with good S/N and fewer aliasing artifacts when compared with standard RTM or standard LSRTM.

Nowadays, Least-Squares Reverse Time Migration (LSRTM) methods are becoming more and more attractive. Despite an obvious computational cost, they are really interesting to improve resolution, reduce migration artifacts and produce better amplitudes. They could give access to amplitude friendly migrated gathers in very complex geological settings. We present here a Least-Squares Reverse Time Migration method using regularization in the surface offset domain. The main objective of this work is to reconcile pre-stack and post-stack LSRTM in order to benefit from the two approaches. Preliminary numerical tests on synthetic datasets will demonstrate the effectiveness of this inversion in comparison with standard LSRTM or RTM. Presentation Date: Wednesday, September 27, 2017 Start Time: 2:15 PM Location: 361A Presentation Type: ORAL

We present a least-squares reverse time migration (LSRTM) method using Radon preconditioning to regularize noisy or severely undersampled data. A high resolution local radon transform is used as a change of basis for the reflectivity and sparseness constraints are applied to the inverted reflectivity in the transform domain. This reflects the prior that for each location of the subsurface the number of geological dips is limited. The forward and the adjoint mapping of the reflectivity to the local Radon domain and back are done through 3D Fourier-based discrete Radon transform operators. The sparseness is enforced by applying weights to the Radon domain components which either vary with the amplitudes of the local dips or are thresholded at given quantiles. Numerical tests on synthetic and field data validate the effectiveness of the proposed approach in producing images with improved SNR and reduced aliasing artifacts when compared with standard RTM or LSRTM. Presentation Date: Tuesday, October 18, 2016 Start Time: 8:50:00 AM Location: 171/173 Presentation Type: ORAL

Stacking is of paramount importance in seismic processing to improve the signal to-noise ratio (S/N) and the imaging quality of seismic data. The conventional stacking method that averages equally a collection of input traces cannot robustly suppress coherent noises. To attenuate this kind of noise and achieve an optimally stacked image remains an attractive and challenging topic in the seismic industry. The key point for ongoing research is to develop methods that can be used to reliably discriminate between “good” and “bad” data samples. To this end it is important to identify two key objectives for the process of optimal stacking. The first is to find a suitable domain where signal can be easily distinguished from noise and the second is to build a robust procedure that allows only the signal contribute to the final migrated stack. We describe a novel, iterative method to clean and enhance the stacked migration image. For bandlimited migration algorithms we define an original prestack image domain that is analogous to the aperture partitioned migration domain of Kirchhoff-type migration techniques. Through an automatic procedure that is based on a coherence analysis we show how, in this domain, signal can be separated from both coherent and incoherent noise in an effective way. With the aid of synthetic examples we show how this yields to a superior quality image compared to a conventional migration stack.