We N Fracture and Stress from in Situ, Full-azimuth Seismic Data - Application to a Kuwait Oil Field Dataset

Transcription

1 1-4 June 2015 IFEMA Madrid We N Fracture and Stress from in Situ, Full-azimuth Seismic Data - Application to a Kuwait Oil Field Dataset B.R. de Ribet* (Paradigm), S.P. Shiv Pujan Singh (Paradigm), D.D. Duane Dopkin (Paradigm), P.K. Pramod Kumar (Paradigm), O.B. Osamah Al Bannaw (Paradigm), N.R. Narahi Srinivasa Rao (Kuwait Oil Company), O.O. Olusegun Kolawole Osuntola (Kuwait Oil Company), S.A. Samar Al Ashwak (Kuwait Oil Company) & Q.D. Qasem Dashti (Kuwait Oil Company) SUMMARY To better understand fractured reservoir behavior, geophysicists often rely on directional groupings of surface recorded seismic data that can be imaged and run through inversion processes to obtain fracture orientation and intensity attributes. We propose an innovative seismic imaging and inversion system carried out in depth, in which seismic data events are imaged and decomposed in the Local Angle Domain into two complementary full-azimuth angle gather systems with fully sampled directivities and reflectivities. The combination of the two angle gathers, together with the ability to handle the full-azimuth information in a continuous manner, enables the generation and extraction of high-resolution information about subsurface angle dependent reflectivity in localized 3D space. It expands our knowledge about both continuous structural surfaces and discontinuous objects, such as faults and small-scale fractures, leading to accurate, high-resolution, highcertainty, velocity model determination and seismic reservoir characterization. The objective of this paper is to describe the technology and to illustrate its benefits by applying it to a set of fractured reservoirs (carbonates and gas shales) in a Kuwait oil field.

2 Introduction Unlike conventional ray-based imaging methods, which initiate rays from the acquisition surface, ray tracing in the local angle domain is performed from the image points up to the surface where diffracted rays are initiated in all angles and directions (including turning rays), forming a system for mapping (binning) the recorded surface seismic data into angle gathers containing full azimuth reflectivity and directivity data. Based on these developments, we propose an innovative seismic imaging system carried out in depth, in which imaged data events are decomposed in the Local Angle Domain (LAD) into two complementary full-azimuth angle gathers: Directional and Reflection. Their combination, together with the ability to handle the full-azimuth information in a continuous manner, enables the generation and extraction of high-resolution information about in situ and localized angle dependent reflectivity. The complete set of information from both angle gather types expands our knowledge about both continuous structural surfaces and discontinuous objects, such as faults and small-scale fractures (Nicolaevich et al., 2013), leading to accurate, high-resolution, high-certainty, velocity model determination and seismic reservoir characterization. This paper describes the technology and illustrates its benefits when applying it to a set of fractured reservoirs (carbonates and gas shales) in a Kuwait oil field. Method Full-azimuth reflection angle gathers True amplitude, full-azimuth reflection angle gathers organize and display seismic reflectivity as a function of the in situ opening angle and opening azimuth. These gathers are most meaningful in the vicinity of actual local reflecting surfaces, where the reflection angles are measured with respect to the derived background specular direction. The reflection angle gathers are used for automatic picking of full-azimuth angle domain Residual Moveouts (RMO) and together with the derived background directivities, provide a complete set of input data for full-azimuth tomography updates to the isotropic and anisotropic velocity models. The full-azimuth angle dependent amplitude variations are used for reliable and accurate residual moveout (RMOZ) and amplitude (AVAZ) inversions to determine fracture and stress orientation and intensity. An example of the rich set of information obtained from a full azimuth 3D reflection angle gather from the project is shown in Figure 1. The gather is displayed as a cylinder with transparency in order to better appreciate the full range of data recovered in the imaging and decomposition process. This full-azimuth gathers image provides diagnostic quality control for the accuracy of anisotropic velocity models (VTI, TTI, HTI), and enables the automatic detection of residual moveout (RMO) errors. The gathers can also be displayed in full-angle or full-azimuth sectors to better understand the influence of azimuth on the velocity model, and to better understand the behavior of seismic amplitude as a function of opening angle and azimuth.

3 Figure 1 3D reflection angle gather as full 3D volume with transparency mode, emphasizing the high reflectivity values associated with a selected azimuthal sector. Two dimensional displays of full-azimuth angle domain gathers are also useful for analyzing velocity or amplitude anisotropy. Figure 2 shows five extracted opening angle panels (10, 15, 20, 25 and 30 degrees). Each panel includes reflections from all azimuths. The azimuthally varying reflector moveout indicates an anisotropic effect, and is marked by an ellipse on the 15 degree opening angle panel. 10 o 15 o 20 o 25 o 30 o Figure 2 Five extracted opening angle sectors, where the full azimuth moveout reflectivity highlighted in the ellipse on the 15 degree opening angle exhibit HTI anisotropy. Directional gather-based imaging Full-azimuth directional angle gathers are constructed from scattered seismic energy decomposed and organized into dip/azimuth angle bins at subsurface points. They contain information from both specular and diffraction energy sources. Specular energy originates from continuous interfaces while diffraction energy originates from discontinuous surfaces and contains information about local heterogeneities. The ability to decompose the specular and diffraction energy from the total scattered field is one of the core components of this imaging system. Weighting seismic data with the specular energy enhances the continuity of the imaged reflectors. The specular energy also provides accurate information about the orientation (dip/azimuth) and continuity of the local reflectors. Conversely, by weighting the seismic data with diffraction energy, it is possible to emphasize geometrical features (e.g. small faults, fractures, reefs, small channels, etc.) which are often lost in the traditional imaging and stacking procedure. Directional angle gathers also provide a convenient decomposition domain for separating primary and multiples energy. Dips and azimuths of even subtle inter-bed multiples can often be differentiated and selectively isolated in directional angle gathers (Figure 3).

4 Figure 3 The image on the left includes the fully decomposed dataset with primaries and multiples. The image on the right shows selective removal of interbed multiples. Our approach consists of three main stages: Ray tracing, full-azimuth angle domain decomposition, and final imaging (weighted stacks). The ray tracing stage involves shooting a fan of rays from image points up to the surface. Ray attributes, such as traveltime, ray coordinates, slowness vectors, amplitude and phase factors, are stored for each ray. The full-azimuth angle domain decomposition stage involves a combination of ray pairs indicating incident and reflected/diffracted rays. Each ray pair maps a specific seismic data event recorded on the acquisition surface, into a four-dimensional local angle domain space in the subsurface - dip and azimuth of the ray pair normal, opening angle and opening azimuth (Figure 4). Figure 4 Surface-to-subsurface and subsurface-to-surface mapping. Results In this study full-azimuth amplitude (AVAZ) inversion was run for the target (Deep Carbonate reservoir, m). The fracture density volume and the volume of azimuth of axis of symmetry (minimum stress) were generated and property maps extracted along the reference horizons for comparison with the full-azimuth residual moveout (RMOZ) inversion. The most important results obtained in this project were the identification of fracture density and fracture azimuthal orientation within the targeted Deep Carbonate reservoir. Figure 5 shows the maps obtained from the full-azimuth inversions in the target areas. On the left is a fracture intensity map. The blue-red color indicates areas of high fracture density. It is superimposed with vectors that display the minimum stress orientation (perpendicular to the fracture orientation) and intensity (defined by the vector length). On the right, a map of fracture density obtained from an AVAZ inversion.

5 Figure 5 Top Deep Carbonate reservoir fracture maps generated from full-azimuth RMOZ inversion (left) and full azimuth AVAZ inversion (right). Conclusion The application of full-azimuth imaging and inversion technology to the Kuwait deep carbonate field produced high-resolution, fit-for-purpose images that emphasized specular and diffracted energy and suppressed challenging interbed multiples. The results obtained from the independent but complementary full-azimuth inversion solutions provided essential information about the distribution and intensities of fracture properties. This information ensured optimal horizontal drilling as well as the correct placement of exploration wells. It was associated with controlled risk and cost reduction by analyzing in situ seismic measurements related to fracture orientation and intensity. Acknowledgements The authors sincerely thank the Management of Kuwait Oil Company for granting permission to present this paper. Permission granted by the Ministry of Oil, the State of Kuwait for publishing this paper is also gratefully acknowledged. References Nicolaevich, I.A., Viktorovich, S.I., Vasilyevich, G.A. and Koren Z. [2013] Applying full-azimuth angle domain pre-stack migration and AVAZ inversion to study fractures in carbonate reservoirs in the Russian Middle Volga region. First Break, 31,