Heavy oil reservoirs are typically not possible to characterize or monitor using seismic methods, since the key acoustic properties of heavy oil namely density and bulk modulus are too similar to those of water. Multi-transient EM (MTEM) is a time-domain dipole – dipole galvanic resistivity mapping technology ideally suited to estimate in situ reserves of heavy oil as well as monitoring of cold and thermal production of heavy oil. All aspects of MTEM data acquisition have been optimized to achieve a sufficiently dense surface sampling in relation to the target depth, and to reach the necessary signal to noise (S/N) at target depth in the shortest time possible with a given source strength. The spatial resolution is sufficient to characterize the total recoverable reserves within a planned Steam Assisted Gravity Drainage (SAGD) project, or within a volume affected by Cyclic Steam Stimulation (CSS), also known as "huff and puff' production. Repeatability of EM data is excellent facilitating monitoring of subtle resistivity changes with good spatial resolution.
Several factors affect the resistivity during the monitoring phase. The increased temperature will lower the resistivity by up to 85 % and the changes in resistivity due to changes in water saturation are described by the Archie equation (Archie, 1942). The salinity of the pore water will be diluted when steam condenses to water resulting in a resistivity increase. In addition to these factors, the reservoir also suffers dissolution of grain cement and loss of some of the original mineral volume. At the same time the reservoir expands due to the increased fluid pressure and concomitant reduction in vertical effective stress. This leads to further increases in porosity and permeability, hence also to decreased resistivity and weakening of the bulk and shear moduli affecting the seismic response.
Ideally a feasibility study for EM monitoring should be performed based on the results generated by a reservoir simulator, where the dynamic changes in the key parameters: temperature, salinity and the saturations of water, oil, hydrocarbon gas and live steam are tracked for every grid cell over time. This information can then be transformed into resistivity for each grid cell and the EM response from the steam chamber can be modeled at different time steps throughout the planned lifespan of the project. Currently, reservoir simulators cannot incorporate dynamic changes in porosity and permeability that are observed as permanent reductions of the elastic moduli and also reduced resistivity. It is necessary to better understand this so called formation damage to fully describe the evolution of the steam chamber.