--> --> Workflow for Zoeppritz AVO Inversion to Estimate Seismic Anisotropy and Geomechanical Properties of Shale: Case Study of Avalon Shale, Delaware Basin
[First Hit]

2019 AAPG Annual Convention and Exhibition:

Datapages, Inc.Print this page

Workflow for Zoeppritz AVO Inversion to Estimate Previous HitSeismicNext Hit Previous HitAnisotropyNext Hit and Geomechanical Properties of Shale: Case Study of Avalon Shale, Delaware Basin

Abstract

Reliable estimation of geomechanical properties (i.e., Young's modulus and Poisson's ratio) of shale provides important constraints to guide production by identifying sweet spots and locations for effective hydraulic fracturing. However, most conventional amplitude variation with offset (AVO) inversions cannot properly estimate the properties, since the inversions are developed based on linear approximations of Zoeppritz equations assuming weak contrasts and Previous HitseismicNext Hit isotropy. Organic-rich shale formations are, however, often anisotropic. In order to overcome limitations of the conventional methods, we propose a workflow to estimate Previous HitseismicNext Hit Previous HitanisotropyNext Hit and geomechanical properties of organic-rich shale. It is based on analyses of an AVO inversion with full Zoeppritz solutions mainly for P-wave reflection amplitudes. The Previous HitanisotropyNext Hit of the model shale is related to the kerogen volume fraction values using measured well logs and laboratory data for Avalon shale, Delaware Basin. By applying inversion tests, we determine behaviors of the AVO inversion solutions developed for isotropic media when the target shale formation instead has Previous HitseismicNext Hit Previous HitanisotropyNext Hit related to organic content. These tests show that the inversion accurately determines horizontal P-wave and S-wave velocities and underestimates density when a far angle range is applied with input data. When the angle range is small, the inversion can obtain reliable vertical velocities, and correct density. Therefore, Previous HitseismicNext Hit Previous HitanisotropyNext Hit of the model can be estimated by comparing these inverted horizontal and vertical velocities. In addition, geomechanical properties of the model are also reliably determined in both horizontal and vertical directions. We also show that addition of converted PS-reflection information improves the inversion results. When we jointly apply PP- and PS-reflections as input data, Previous HitseismicNext Hit Previous HitanisotropyTop and geomechanical properties are more reliably estimated than PP AVO inversion for both noise-free and noise-introduced tests.