Integrated Marcellus Shale Characterization Using Multicomponent Seismic Data in Northern West Virginia
We integrate seismic, geologic and petrophysical data to characterize the Marcellus Shale in northern West Virginia. We focus on several key factors that affect shale reservoir exploration and development. First, seismic data provides important structural and stratigraphic information for planning and drilling horizontal shale oil and gas wells. Secondly, with P wave and S wave anisotropy estimation from full azimuth, multi-component seismic data, we can characterize stress orientation and magnitude, which is also important for well planning and hydraulic fracturing design. Finally, we use joint prestack PP and PS wave inversion to estimate the P and S wave impedance and density, and further compute the geomechanical rock properties and facies. Complex structural styles are observed in the Ordovician and Devonian stratigraphic intervals using multicomponent seismic data. The Silurian Salina Salt movement created major through cutting faults in this area, though the fault surface dip changes due to mechanical stratigraphy. We also observed many faults that are limited to the Devonian layer above the Onondaga limestone. In some cases multicomponent PS data defines the fault offset much clearer than PP data. We also observed strike slip faults and major reverse faults in our survey. Anisotropy can indicate fracture intensity or horizontal stress differences. Azimuthal P wave fast velocity seems follow J1, J2 and J3 fracture azimuths. Windowed azimuthal AVO analysis shows higher resolution anisotropy azimuth and magnitude variability. Converted wave (PS) fast shear azimuth agrees with the dominate fault strike. Seismic attributes including coherence, fault probability and curvatures are used together with the multiple anisotropy volumes for interpretation of fracture and stress in the Ordovician and Devonian stratigraphic zones. We demonstrate that converted PS wave data is of high quality. Joint prestack inversion of PP and PS data produce stable P and S impedance and density distribution with stratigraphic zones for reservoir characterization and geomechanical applications. We use Bayesian probabilistic methods to estimate petrophysical properties, such as TOC content and lithofacies within the Marcellus shale reservoir.
AAPG Datapages/Search and Discovery Article #90189 © 2014 AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 6–9, 2014