Fracture characterization of a carbonate reservoir in the Arabian Peninsula

Estimation of reservoir fracture parameters, fracture orientation and density, from seismic data is often difficult because of one important question: Is observed anisotropy caused by the reservoir interval or by the effect of the lithologic unit or multiple units above the reservoir? Often hydrocarbon reservoirs represent a small portion of the seismic section, and reservoir anisotropic parameter inversion can be easily obscured by the presence of an anisotropic overburden. In this study, I show examples where we can clearly observe imprints of overburden anisotropic layers on the seismic response of the target zone. Then I present a simple method to remove the effect of anisotropic overburden to recover reservoir fracture parameters. It involves analyzing amplitude variation with offset and azimuth (AVOA) for the top of reservoir reflector and for a reflector below the reservoir. Seismic CMP gathers are transformed to delay-time vs. slowness (tau-p) domain. We then calculate the ratio of the amplitudes of reflections at the reservoir top and from the reflector beneath the reservoir. The ratios of these amplitudes are then used to isolate the effect of the reservoir interval and remove the transmission effect of the overburden. The methodology is tested on two sets of models - one containing a fractured reservoir with isotropic overburden and the other containing a fractured reservoir with anisotropic overburden. Conventional analysis in the x-t domain indicates that the anisotropic overburden has completely obscured the anisotropic signature of the reservoir zone. When the new methodology is applied, the overburden effect is significantly reduced. The methodology is also applied to an actual PP surface reflection (Rpp) 3D dataset over a reservoir in the Arabian Peninsula. Ellipse-fitting technique was applied to invert for two Fracture parameters: (1) Fracture density and (2) fracture direction. Fracture density inversion results indicate increased fracturing in the anticline structure hinge zone. Fracture orientation inversion results agree with Formation MicroImaging (FMI) borehole logs showing a WNW-ESE trend. This newly developed amplitude ratio method is suitable for quantitative estimation of fracture parameters including normal and tangential “weaknesses” (ΔN and ΔT respectively). Initially, inversion of conventional AVOA for ΔN and ΔT parameters indicates that the ΔN parameter is reliably estimated given an accurate background isotropic parameter estimation derived from borehole logging data. While ΔN parameter inversion is successful, inversion for ΔT parameter from Rpp information is not, presumably due to the dependence of ΔT estimation on many medium parameters for accurate prediction. The ΔN parameter is then successfully recovered when applied to the amplitude ratio values derived from synthetic data. It is important to recognize that ΔN parameter is directly proportional to fracture density and high ΔN values can be attributed to high crack density values. The ΔN parameter inversion is also applied to the amplitude ratios derived from real seismic data. This inversion requires fracture azimuth data input that is obtained from the fracture direction inversion using ellipse-fitting technique. The background Vp/Vs ratio.