Most of the oil produced nowadays comes from aging fields where the rate at which it is replaced by new field discoveries has been declining in recent decades. Conventional light oil is becoming harder to find so companies are looking to exploit huge resources of more viscous oil. Advances in drilling technology, increased computational power and improved reservoir management made it easier and cheaper to produce more oil from the vast reserves of heavy oil worldwide. Steam injection is the most widely used enhanced oil recovery method for these oils. Better monitoring of the steam chamber in a steam injection operation helps to increase the sweep efficiency and improve the management of the reservoir.
This thesis investigates the feasibility for using AVO to better monitor steam injection in heavy oil reservoirs. Many factors such as temperature, pressure, fluid saturations and oil viscosity interact during thermal production of heavy oil, which complicates the interpretation of seismic data.
The investigation is based upon a geological model inspired by a real field case of Emlichheim oil field in Germany. The research uses a combination of reservoir simulation, rock physics and amplitude versus offset seismic analysis to investigate the characterization of the vertical extent of the steam chamber using AVO. This is done by first using a reservoir simulation software (STARS) to model the dynamics of steam injection into a heavy oil reservoir with properties similar to those of Emlichheim. The reservoir properties obtained from the simulation are then used as input into a rock physics code that was developed in MATLAB. This code calculates the P- and S-wave velocities along with the densities of the saturated rock. This is then used as input into HampsonRussell software, which calculates the intercept and gradient of the two-term Aki-Richards approximation of the amplitude versus offset (AVO) equation. A sensitivity study was conducted to test the effect of temperature, porosity, pressure, steam saturation and oil API on the intercept and gradient of the AVO approximation. Pre steam injection, the AVO gradient and intercept are most sensitive to porosity and pressure changes at the top of the reservoir. Post steam injection, the AVO intercept and gradient are most sensitive to porosity, pressure and steam saturation at the top of the reservoir, while at the bottom of the steam chamber they are most sensitive to steam saturation, porosity and pressure.
In this thesis we study factors that influence steam flood performance and sensitivity of the AVO response to various possible rock and fluid properties. The study of the gradient and intercept of the AVO at the base of the steam chamber concluded that they are both positive at the base of the steam chamber within the reservoir. This led to a novel technique in estimating the bottom of the steam chamber, thus estimating the thickness of the steam chamber. This is a very important and powerful tool as it allows us to be more confident in our pick of the bottom of the steam chamber when it is not possible to visually identify the bottom of the steam chamber using time-lapse seismic surveys. This helps to delineate the vertical extent of the steam front, which can be used for history matching and improving the reservoir simulation which ultimately helps increase the sweep efficiency and the oil recovery factor by better managing the oil reservoir.