Fluid Discrimination in Ultra-Deep Reservoirs Based on a Double Double-Porosity Theory

We develop a methodology, based on rock-physics templates, to effectively identify reservoir fluids in ultra-deep reservoirs, where the poroelasticity model is based on the double double-porosity theory. P-wave attenuation, the ratio of the first Lame constant to mass density (lambda/rho) and Poisson ratio are used to build the templates at the ultrasonic and seismic frequency bands to quantitatively predict the total and crack (soft) porosities and oil saturation. Attenuation on these frequency bands is estimated with the spectral-ratio and frequency-shift methods. We apply the methodology to fault-controlled karst reservoirs in the Tarim Basin (China), which contain ultra-deep hydrocarbon resources with a diverse pore-crack system, low porosity/permeability and complex oil-water spatial distributions. The results are consistent with well-log data and actual oil recovery. Crack porosity can be used as an indicator to find regions with high oil saturation, since high values implies a good pore connectivity.