TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractThe determination of reservoir saturation using the seismic response is an important element of predicting the performance of fractured reservoirs. One of the primary saturation problems the seismic signal can address is the location of the gas cap with respect to the water or oil leg. The classical Gassmann (1951) approach to predicting the effects of saturation upon the seismic signal is not valid for fractured reservoirs. Instead, the results of Brown and Korringa (1975) that account for the anisotropy of the dry rock are used to predict the effects of saturation upon the seismic signals. For a single set of vertical, aligned fractures the Pwaves are most sensitive to the saturation when they are nearly perpendicular to the fractures. For S-waves, a classical result is obtained: the S-waves are not sensitive to the saturation within the vertically aligned fractures. However, when the elastic symmetry of the dry rock differs from the case of vertical fractures, the S-waves become sensitive to the saturation. This can happen when either the fractures are tilted slightly away from the vertical or when the fractures are best described as rough surfaces. In these cases, S-waves are most sensitive to the saturation when propagating parallel to the cracks with a polarity perpendicular to the cracks. These results indicate that an S-wave "bright spot" or region of increased S-wave splitting can be used to predict gas versus brine or oil-saturated portions of the reservoir. Since P-waves have the maximum sensitivity to the saturation of a fractured reservoir when they propagate perpendicular to the fractures, they require long-offset surface seismic methods that may be difficult to implement in practice. However, since vertically propagating S-waves with different polarizations can be used to study the effects of the saturation, they offer the most potential for mapping saturation within fractured reservoirs.

TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractThe permeability and seismic properties of fractured reservoirs are not easily correlated. A part of the problem is that seismic properties are controlled by the elastic properties of the rock grains, the pore-space/crack-space geometry, the connectivity of the pore spaces and any local flow that occurs during the transmission of the seismic signal. Permeability, however, is governed by the pore-space geometry, the effective length and apertures of the cracks as well at the connectivity of the cracks. One way to calibrate the permeability and the seismic is to identify a "sweet spot" within a reservoir where certain simplified assumptions can be made regarding the fracture permeability. Under these ideal conditions, it is possible to correlate the seismic response and the production data using simplifying assumptions.After the calibration is accomplished at the well control, these results can then be used to predict the permeability variations away from the well control using the seismic response and models for the relationship between the permeability and the seismic properties.This approach requires the development of empirical relations between the seismic response and the permeability of the reservoir.