Search

The source-over-streamer configuration has been designed to improve pre-processing of the seismic data leading to high-resolution imaging. It also leads to new residual moveout information, unique in its high quality and density. By adding a front source to the original design, long offset information is now available for full waveform inversion velocity update. The combination the reflection information from the clean common imaged gather and the diving wave from the long offset allows a high-resolution anisotropic model to be obtained.

We have implemented 3D faults as discontinuity surfaces, of finite extent, in the RLM-3D inversion regularization, and used the scheme during both 3D cooperative and cross-gradient joint inversions of geothermal MT and gravity data, firstly for synthetic model, and then for the Sorik Marapi graben. Through integrated, quantitative modeling of multiple geophysical data types over geothermal fields, now including faults as sharp discontinuities, we facilitate geologically and structurally reliable multi-property 3D earth models that consistently explain the observations of all geophysical dataset.

Recent years have seen growing interest in improved shallow resolution images of the subsurface. This has led to ever more innovative acquisition approaches, each tailored to individual geological settings. We focus on a towed streamer acquisition in the Barents Sea which deployed five sources above the streamers for high definition imaging, along with a single source towed behind the streamer vessel to acquire long offset data for full waveform inversion. Firstly we demonstrate how accurate deblending is a key processing step to uncover the potential of these data.

Deep convolutional neural networks (DCNNs) are growing their popularity in seismic data processing due to their previous achievements in signal and image processing. In this paper, we explore the link between DCNN and seismic processing. We use two examples to demonstrate the potential of the application of DCNNs to seismic processing. More importantly, we discuss challenges and issues to solve before deploying DCNNs to production, and suggest some directions of study.

This work investigates the potential of QEMSCAN in assessing the pore structure system. 23 carbonate and clastic samples are analysed using QEMSCAN, which shows a good assessment of the macropore system (pore larger than 10µm). It defines the macropore system volume and delivers pore length distribution. The pore length distributions are integrated with plug-NMR results; a good match is observed between the two type of results, which enables a calibration of the T2 relaxation time with pore sizes. Compared to NMR data, QEMSCAN provide real pore measurements, which could be more accurate to use in permeability calculations.

Using a dense North Sea 4D dataset over the Golden Eagle field, we demonstrate the impact of OBN density on both the 3D and 4D seismic image by migrating progressively sparser node configurations. It is shown that 4D image quality is more sensitive to changes in node density than is the 3D image, and an acceptable sparse survey for 3D imaging may be inadequate for 4D. Mitigating the effects of reduced node density via processing shows partial success, but the tests highlights the importance of node density in 4D survey design for North Sea OBN data.

Challenged by the presence of strong anisotropy and velocity reversal in the near-surface, we apply structurally constrained anisotropic Multi-Wave Inversion (MWI), over 1200 km2 of OBC data offshore Abu Dhabi. MWI aims at simultaneously inverting the P-wave first breaks (FB), the ground roll Dispersion Curves (DC) and the Vertical Two-Way Times (VTWT) of main interfaces. In this study, MWI is extended to invert for epsilon that is constrained by the FB and the VTWT. We highlight the importance of proper preconditioning of the DC and FB inputs using data mining tools when applying the method to a large production dataset.

This paper shows how to calculate the bias due to misspecified models in least-squares parameter estimation. It introduces Omitted Variable Bias (OVB), a technique well known in least-squares analysis in the context of econometric data analysis. OVB is applied to the analysis of linearized isotropic AVO models, both analytically and numerically. For misspecified models, such as two-term AVO fitting with large angle range or with large contrasts, OVB provides relations between the biased and unbiased least-squares model parameters. A Jupyter notebook and binder link let’s the reader apply the ideas presented to their own AVO models.

An article summarizing CGG's recent Multi-Client seismic acquisition in the West of Shetland. Article covers the exploration history, key petroleum systems, geological challenges for seismic imaging and interpretation and identification of exploration opportunities within the data.