Seismic measurements acquired at different stages in the life of a reservoir can monitor the fluid distribution over production time. Seismic monitoring has the potential to significantly enhance recovery and optimize exploitation schemes in existing and new fields. The production from the reservoir induces changes in saturation, pore pressure and stresses, which may influence the process of wave propagation in rocks. The influence of mean effective stresses, due to changes in mean total stresses or/and pore pressure on reservoir elastic properties are not always taken into account in a proper manner. The proposed methodology evaluates what is the contribution of a geomechanical approach on the computation of elastic parameters. The implementation of this method is composed of several steps. First, in order to take into account multiphase fluid flow, pressure and saturation are computed through reservoir simulation. Then, the computed pressure is used as a load in the geomechanical modeling. Using poroelasticity theory introduced by Biot, the contribution of rock deformation to fluid flow is evaluated. This allows the simulation of the stresses and pressure distributions throughout the whole geological structure over production time. The following step consists in generating elastic parameters as function of reservoir effects using rock physics. In order to take into account the mean effective stresses on seismic velocities, the Hertz-Mindlin's contact theory is used. The saturation effect on seismic velocities is then considered by Biot-Gassmann's relation. This methodology has been validated on real repeated data for monitoring an underground gas storage. This integrated procedure is then applied to several scenarii of production. Thus, the sensitivity of elastic parameters has been analyzed in order to differentiate the different reservoir effects. Then, using elastic parameters, seismic modeling allows the generation of seismic responses at well location that reveal the patterns of expected seismic monitoring results. Some seismic attributes, such as time-shift delay have been measured. They show that careful processing of seismic data is required for seismic monitoring. This integrated procedure applied on real data for monitoring an underground gas storage leads to demonstrate the play of differentiated attributes involving P and S-waves to separate mean effective stresses effects from saturation ones. Using elastic modeling, the impact of offset changes was demonstrated to measure reliable time-lapse parameters like time-shift delay attribute and amplitude variations. Seismic measurements acquired at different stages in the life of a reservoir can monitor the fluid
展开▼
