No abstract

Optimization of production from unconventional reservoirs requires estimates of reservoir properties such as porosity, total organic carbon (TOC) content, clay content, fluid saturation, and fracture intensity. The porosity and TOC content help to determine reservoir quality, and the natural fracture intensity provides information important for the completion strategy. Because shale reservoirs display intrinsic anisotropy due to layering and the partial alignment of clay minerals and kerogen with the bedding plane, the minimum acceptable representation of the anisotropy of naturally fractured shale-gas reservoirs is orthotropy, in which a set of vertical compliant fractures is embedded in a vertical transverse isotropic (VTI) background medium. Full-azimuth seismic data are required to characterize such reservoirs and to invert for the anisotropic elastic properties. Orthotropic inversion uses azimuthally sectored seismic data stacked according to the incident angle. Even for high-fold acquisition, this azimuth/angle grouping can result in low-fold angle stacks. Orthotropic amplitude-variation-with-offset-and-azimuth (AVOAz) inversion requires seismic preconditioning techniques that ensure proper primary amplitude preservation, noise attenuation, and data alignment, and a workflow implemented for the construction of an orthotropic rock-physics model. This model integrates well and core data to estimate reservoir properties using the results of the AVOAz inversion. The seismic inversion results include the P- and S-impedance and parameters quantifying the azimuthal anisotropy. The rock model assumes a VTI kerogen-rich layer, containing aligned vertical fractures, and it uses prestack orthotropic AVOAz inversion results to predict porosity, TOC, and fracture intensity.

A 2D seismic reflection line for shallow targets in north Kuwait was acquired in December of 2010 with varying source types: explosives, accelerated weight drop, small (15,000 lb) vibrator, and heavy (80,000 lb) vibrator. The exploration target is shallow at around 100 ms to 400 ms two-way time (TWT). The objective of the study was to establish the optimal source effort for a planned 3D survey. The accelerated weight drop proved to be inferior to the other source types because the data lacked high-frequency content. Explosive data benefit from the absence of engine noise and less surface-wave noise compared to the other source types used; yet, the overall energy penetration was not as good as expected. The best data were achieved with vibrators. The small 15,000 lb vibrator produced less engine noise, but fell somewhat short with the observed frequency bandwidth compared to the heavy 80,000 lb vibrator because of vibrator design and energy penetration. The final stacked sections of the two vibrator data are very comparable, however, in the overall imaging of the subsurface. In addition, we acquired the 2D line a second time with the heavy vibrator refitted with an exhaust silencer kit that reduced the engine noise and air wave in prestack data. This has improved the signal-to-noise (S/N) ratio in the stacked data by 2 dB.

No abstract

No abstract