A feasibility study of land CSEM reservoir monitoring in a complex 3-D model

Abstract

S U M M A R YWe carried out a series of numerical simulations in a complex 3-D resistivity model to investigate the feasibility of using controlled-source electromagnetics on land for monitoring changes in a hydrocarbon reservoir during production. Displacement of oil by saline water injection changes the resistivity. The modelling allows a comparison of the measured timelapse EM signal to various sources of noise that can be expected in a field experiment, for instance, magnetotelluric signals, repeatability errors and near-surface resistivity changes caused by seasonal variations. Our estimates show that land CSEM monitoring should be feasible, though not easily, for the example considered here, a thick reservoir at a depth of about 1 km. The trade-off between signal strength and repeatability errors requires the source to be located at some distance from the reservoir. Measurements in a monitoring well suffer less from surface noise. Measuring the vertical electric component in a well, placed at some distance from the reservoir, provides the best result.

Sensitivity of the near-surface vertical electric field in land controlled-source electromagnetic monitoring

“…Previous feasibility studies on land CSEM monitoring clearly indicate that E z should be the field component most sensitive to resistivity changes at depth. E z measurements in a well at some distance from the reservoir should provide highest sensitivity to relatively small resistive reservoirs Wirianto et al, 2010).…”

“…On land, though, reservoir volumes affected by resistivity change need to be large before they become detectable using standard surface-based horizontal dipole-dipole configurations (Wirianto et al, 2010). Therefore, we investigate the suitability of the vertical electric field for detecting resistivity changes related to steam injection in an onshore setting.…”

“…However, this requires substantial experimental effort; developing less expensive monitoring techniques is thus desirable. The resistivity difference between hydrocarbon-bearing and steam-saturated rocks can be significant, making the CSEM method potentially suitable for reservoir monitoring (Wright et al, 2002), provided that the resistivity change can be detected in the presence of repeatability errors and noise (Wirianto et al, 2010).…”

Focused time-lapse inversion of radio and audio magnetotelluric data

“…Orange et al (2009) and Lien and Mannseth (2008) considered CSEM monitoring for marine applications, while Wirianto et al (2010) presented a feasibility study of land-based CSEM. They all conclude that CSEM monitoring is feasible, though not easy due to the diffusive character of EM signals and the low frequencies required to reach the reservoirs.…”

Green's Tensors for the Diffusive Electric Field in a Vti Half-Space

“…The numerical simulation of these experiments, for scenario studies [24] or full blown inversion [14], requires a dense grid around the source for accurate modeling of the source singularity. In the primary-secondary formulation, the fields are split into a part that is computed analytically and a secondary part that involves the full partial differential equations driven by a source term based in the same equations applied to the primary field; see [25], for instance. This will not change the convergence rate of an iterative solution method for the discretized problem, but generally improves the accuracy of the computed target response.…”