What’s new in exploration

 Vol. 233 No. 10

The use of land seismic data is increasing, from sophisticated 3D programs in the Middle East and North Africa, to interior basins in China and shale plays in North America and Europe. High-definition 3D surveys will lead to future 4D efforts.

In March 2007, the Society of Exploration Geophysicists (SEG) launched the SEG Advanced Modeling Program (SEAM), an industrial consortium dedicated to large-scale, leading-edge geophysical numerical modeling. SEAM is led by a board of nine directors.

According to SEG, SEAM’s mission is to advance the technology of applied geophysics through a cooperative industry effort focused on subsurface model construction and generation of synthetic data sets for geophysical problems of importance to the resource extraction industry. Data sets, after two years of confidentiality, will enter the public domain.

Phase I. SEAM Phase I (2007–2010) addressed subsalt imaging in tertiary basins, with emphasis on the deepwater Gulf of Mexico. The project has 23 industry participants and an industry budget of $2.3 million. This contribution was supplemented with RPSEA funds that enabled an expansion of the earth model to include acoustic simulations with an absorbing upper boundary, tilted transverse anisotropy (TTI) simulations, elastic simulations, gravity response, controlled source electromagnetic (CSEM) response, and magnetotelluric (MT) response. The project was completed earlier this year.

The project results included developing the numerical SEAM earth model, a realistic salt structure within a 60-block area of the Gulf (40 × 35 × 15-km model) with varying rock properties that produce complex imaging. The teams developed a data compression scheme, a benchmark acoustic modeling code and 11 “classic” data subsets for use by industry and academia. SEAM Phase I coding (IWAVE) was provided by the Rice University Inversion Project. Project teams also simulated about 65,000 shot records for an acoustic dataset. The research generated 11 SEAM-related expanded abstracts for the 2011 SEG Annual Meeting, and the final report is available as an eBook from SEG.

Phase II. The focus of SEAM Phase II is to address critical land seismic challenges through advanced 3D elastic modeling. Three core themes were identified during 2009–2010:

• High-density and areal-extensive acquisition geometries
• Near-surface issues
• Fracture detection and fractured reservoir characterization.

SEAM Phase II development began in early 2011, managed by Dr. Michael L. Oristaglio, a senior scientist at Yale University.

Sponsoring firmsprovide funding and technical expertise over Phase II’s three years. Participants include Anadarko Petroleum, CNPC/BGP, BHP Billiton, BP, CGGVeritas, Chevron, ENI, ExxonMobil Upstream Research , Global Geophysical, HESS, ION (GX Technology), Marathon, Oxy, Repsol, Saudi Aramco, Shell, Sinopec, Statoil, Talisman Energy, Total, Tullow Oil and WesternGeco.

John Doherty, principal geophysicist with Tullow Oil, which joined Phase II in June 2012, said, “Becoming a member of SEAM fits within our strategy of accessing relevant leading edge geophysical research through collaboration. This industry-led consortium is studying the geophysical response to realistic geological subsurface models on a scale, and with a degree of realistic complexity which we, like many other participants, could not construct independently.”

Vendors for SEAM Phase II (2012–2013) are providing land-seismic modeling and production simulations for this large, elastic-wave modeling project that includes topography, anisotropy and attenuation. Midland Valley Exploration was added as a Phase II vendor in September, to merge two partial geologic models using its proprietary Move software.

Models. According to SEG, SEAM Phase I attacked the computational challenge in stages, starting with a purely acoustic-wave model with variable density, and proceeding through a quasi-acoustic tilted transversely isotropic (TTI) model to an isotropic elastic model.

The plan for Phase II was to start with isotropic elastic modeling and add the complexities of attenuation (visco-elasticity) and general anisotropy. Given a range of models with different sizes and features to compute over the course of the project, one of the key technical goals is to identify numerical algorithms best suited to each model or geologic feature.

The technical committees in Phase II are investigating high-accuracy approaches to elastic modeling, such as spectral finite-element and discontinuous Galerkin methods, and will consider multi-grid and domain decomposition. These techniques are not part of routine modeling.
 
 Four initial models have been defined and are being built in Phase II:

1. Near-surface arid model (Saudi Aramco, Marathon and BP) is 10 km × 10 km × 500 m  and incorporates a series of features characteristic of desert areas.
2. Near-surface foothills model (Total, Repsol, ExxonMobil and Sinopec) contains extreme topography typical of mountainous thrust zones, with elevation differences and strong lateral and vertical velocity variations at scales comparable to exploration seismic wavelengths.
3. Subsurface unconventional reservoir model (BP and Chevron) is based on the Eagle Ford and Woodford shales, covering 100 sq km, with two types of shale reservoirs.
4. Subsurface thrusted reservoir model (Total and Repsol) will be developed using the Move software.

These SEAM II projects will help the industry make decisions about the future of land-seismic technology.