A New Method of Estimating Water Saturation Using Acoustic Logs – Applications to Oil Reservoirs

Abstract Archie's saturation model is the most widely used empirical model for estimating water saturation in oil and gas reservoirs. The model requires formation resistivity, formation salinity, and electrical properties to accurately estimate water saturation. Archie's model assumes that the conductivity of a reservoir rock is essentially due to the fluid filling the rock's pore space. However, it is well known that in some shaly sand reservoirs and low resistivity low contrast reservoirs, Archie's assumptions are invalidated, and the model does not yield accurate water saturation estimates. Several other saturation models have been developed to address reservoirs where brine salinity is low and or clay content is high. But these other models have their limitations and, in some cases, may not be more effective than Archie's equation. We present a new semi-empirical model that does not require formation resistivity or brine salinity to estimate water saturation and as such is not dependent on the assumptions that conventional saturation models rely on and the limitations that result from those assumptions. The new model instead uses measured compressional and shear sonic logs. The model is based on the lower bulk modulus observed in oil and gas reservoirs. It is shown to be consistent with theoretical rock physics models. Unlike conventional saturation models, the new model does not require formation resistivity to estimate fluid saturations. Instead, the model uses the P-wave and S-wave velocities computed from compressional and shear sonic logs, as well as empirical P-wave-S-wave velocity trends. We have applied the model to nine different sandstone reservoirs offshore the Gulf of Mexico and the North slope of Alaska. The seven offshore Gulf of Mexico sand reservoirs from 4 different wells are all black oil reservoirs, with oil gravity ranging from 27° to 35° API. Water saturation estimates from the new model are compared to conventional saturation models and core water saturation. In all 9 cases, the model yields accurate water saturation estimates that are comparable to those from conventional saturation models. The new model complements resistivity-based saturation models in the evaluation of conventional reservoirs where Archie's model assumptions apply. However, when it comes to low resistivity low contrast reservoirs, the model would yield more accurate estimates of water saturation as evident from the brine salinity sensitivity analysis we have conducted.