Seismic texture analysis for reservoir prediction and characterization

A plethora of seismic attributes are currently in use for reservoir prediction and characterization. Some are useful in understanding subsurface stratigraphy (like channels) while others are useful for structural interpretation (anticlines, faults/fractures etc). Volumetric computation of seismic attributes is helpful to interpreters for 3D seismic visualization and interpretation. In addition to the time of a seismic reflection (resulting in a time-structure map), the more commonly used seismic attributes are rms amplitude, frequency, coherence, AVO, and impedance. These attributes are based on clearly established morphological or petrophysical models with some attributes sensitive to rock types, some to fluid saturation, some to porosity and some to minor faults or fractures within the reservoir. Our case study of a diagenetically altered Mississippian limestone resulting in complex paleo topography defies simple analysis. Even with adequate well control and good ties to 3D seismic data, uncertainty in the attribute expression of different reservoir architecture may result in the failure to identify sweet spots for drilling.