Reservoir simulation CREWES Research Report — Volume 16 (2004) 1 Reservoir simulation and geomechanical modeling in Leming Lake, Alberta John J. Zhang, Tony Settari and Laurence R. Bentley ABSTRACT This paper deals with reservoir simulation and geomechanical modeling. The physics of fluid flow and geomechanical systems is expressed as a set of conservation equations and simulation is their numerical solutions. GEOSIM is used to carry out simulation of a coupled reservoir and geomechanical model created using well logs, core measurements and seismic data. Reservoir simulation shows a spatial pattern of pressure, temperature and saturation changes around wells, which can be used to predict the change in seismic response. Geomechanical modeling shows significant displacements in the reservoir and in areas far away from the reservoir, but significant changes in stress and strain fields are mainly limited to the reservoir. Elastic moduli decrease due to exsolved gas released by lowering pressure and high temperature after a period of production. The moduli will be further reduced due to horizontal fractures predicted by geomechanical modeling. The cumulative effect is expected to be strong enough to be detected by time-lapse seismic data. INTRODUCTION Heavy oil has been produced using cyclic steam stimulation from the Clearwater formation in Leming Lake, Alberta, Canada. Typically the process consists of scheduled cycles of injection and production from vertical, deviated or horizontal wells. High injection pressure is used to generate fractures, which sequentially introduces high temperature steam into the reservoir in hopes of improving injectivity and productivity. The steam chamber and temperature and pressure zones are conventionally forecasted with reservoir simulation. But the prediction is far from complete due to heterogeneity. Recently Imperial Oil shot time-lapse three-D seismic surveys over a few production pads in an attempt to monitor fluid flow and reservoir conditions. Seismic interpretations for changes in saturation, pressure and temperature in reservoirs add another constraint to reservoir simulation in addition to production history matching. Reservoir management based on reservoir simulation optimized by both production performance and time-lapse seismic would enhance heavy oil recovery. In this paper the authors focus on reservoir simulation and geomechanical modeling with an aim to evaluate how feasible time-lapse seismic can be in monitoring reservoir recovery. UNDERLYING PRINCIPLES Fluid injection or/and production from a reservoir disturb the static equilibrium of pore fluids and solid framework. A natural tendency to reach a new balance initiates fluid flow and deformation. During the dynamic process conservation laws can be applied to affected. Equation (1) is the mathematical formula for fluid conservation (notation found at end of paper).