Transmission Tomography

Key Takeaways

• Crosswell seismic transmission tomography achieves spatial resolution that is substantially finer than surface seismic reflection surveys in the interwell region, because the high-frequency seismic sources used in downhole tomography (typically 100 to 3,000 Hz, compared to 10 to 100 Hz for surface seismic) enable wavelengths of 1 to 20 feet in typical reservoir rocks, providing vertical and lateral velocity resolution of the same order: the crosswell tomography acquisition geometry involves a piezoelectric or vibroseis borehole seismic source at a sequence of depths in one well that transmits seismic energy across the interwell formation, with a downhole geophone or hydrophone receiver array in the adjacent well recording the transmitted waveform at each source-receiver combination; for a wellbore pair with 500 feet of productive formation and a receiver array of 100 elements, a full crosswell dataset contains on the order of 50,000 to 100,000 source-receiver ray paths that sample the interwell formation at crossing angles, providing the redundancy needed for stable tomographic inversion and the angular coverage needed to separate velocity variations in the vertical and horizontal directions; the inversion of the crosswell travel time dataset (typically using algebraic reconstruction techniques ART, simultaneous iterative reconstruction SIRT, or full-waveform inversion FWI) produces a 2D velocity model in the vertical plane between the two wells that reveals formation heterogeneity including shale baffles, high-porosity zones, fracture swarms, and fluid contact positions that are below the resolution of surface seismic.
• Time-lapse crosswell seismic transmission tomography (4D crosswell tomography) provides a powerful tool for monitoring reservoir fluid substitution during enhanced oil recovery (EOR), CO2 sequestration, and steam flooding operations, because the acoustic velocity of reservoir rocks is sensitive to the fluid type and saturation occupying the pore space: a baseline crosswell tomography survey is acquired before the injection operation begins, establishing the initial velocity distribution across the interwell formation; subsequent monitor surveys acquired at intervals during the injection program (typically quarterly to annually) are differenced against the baseline to produce 4D velocity anomaly maps that show which parts of the formation have experienced fluid substitution; CO2 injection into saline aquifers or depleted oil reservoirs causes a decrease in acoustic velocity (because CO2 is more compressible than brine or oil at reservoir pressure and temperature, lowering the bulk modulus of the fluid-saturated rock by the Gassmann equation), producing a negative velocity anomaly in the CO2-saturated zones that grows spatially with each monitor survey; steam flooding in heavy oil reservoirs produces large positive temperature effects (steam reduces oil viscosity and the temperature increase affects rock elastic properties) that create strong velocity anomalies detectable in crosswell and surface seismic 4D surveys; the inversion of these velocity anomaly patterns into saturation and fluid substitution estimates requires integration with the Gassmann fluid substitution framework and calibration against production data and well logs in the monitoring wells.
• VSP transmission tomography uses multiple offset surface shots recorded by receivers in a borehole to solve for the velocity distribution and anisotropy parameters in the formation surrounding the wellbore, providing a bridge between the high-resolution but depth-limited wireline log measurements and the lower-resolution but spatially extensive surface seismic reflection survey: the VSP geometry with sources at varying offsets from the wellhead generates first-arrival travel times at the downhole receivers from different propagation directions (near-vertical for zero-offset VSP, increasingly off-vertical for far-offset and walkaway VSP sources), and the tomographic inversion of these travel times solves for the 2D or 3D velocity model around the wellbore that minimizes the misfit between computed and measured travel times; the VSP tomography result produces a velocity model that ties the local wireline velocity (from sonic logging) to the regional surface seismic velocity, improving the accuracy of the depth-to-time conversion that is used to place surface seismic reflectors at the correct depths for well planning and reservoir mapping; in formations with intrinsic seismic anisotropy (transverse isotropy from shale layering or orthorhombic symmetry from vertical fractures combined with horizontal layering), the multi-offset VSP travel times contain information about the Thomsen parameters that characterize the velocity anisotropy, and the anisotropic tomographic inversion recovers these parameters simultaneously with the isotropic velocity structure.
• Electromagnetic transmission tomography between boreholes (crosswell EM tomography) operates at frequencies from 10 Hz to 100 kHz (depending on the formation resistivity and interwell distance) and exploits the sensitivity of electromagnetic wave propagation to subsurface resistivity, which in turn reflects the fluid saturation and fluid type in the pore space: the crosswell EM acquisition system consists of a magnetic dipole or electric dipole transmitter lowered to successive depths in one well and an array of EM receivers (measuring magnetic field components or electric potential) at corresponding depths in an adjacent well; the amplitude and phase of the transmitted EM signal at each source-receiver combination carry information about the integrated resistivity along the propagation path, and the tomographic inversion of the multi-path dataset produces a 2D resistivity cross-section between the wells; EM transmission tomography has been successfully applied to monitoring waterflood front advance (the displacing brine has lower resistivity than the displaced oil, creating a resistivity contrast that appears as a moving boundary in time-lapse EM surveys), CO2 flood monitoring (CO2 increases resistivity relative to the brine it displaces), and steam flood monitoring (steam zones are highly resistive compared to cold formation brine); the spatial resolution of crosswell EM tomography is limited by the electromagnetic skin depth at the measurement frequency, which is typically 30 to 300 feet in typical reservoir rocks and is much coarser than the spatial resolution of crosswell seismic tomography, making EM tomography most useful for imaging large-scale fluid boundaries rather than fine-scale lithological heterogeneity.
• Acoustic transmission tomography in near-surface geotechnical and civil engineering applications shares the mathematical framework with oilfield crosswell seismic tomography but operates at higher frequencies and shorter scales, providing an important methodological context for understanding the technique's capabilities and limitations that apply equally at both scales: in geotechnical applications, borehole-to-borehole acoustic tomography is used to image foundation conditions between adjacent boreholes at construction sites, to locate voids and anomalous zones in karst terrain, and to monitor compaction grouting effectiveness in embankment dams; the fundamental trade-off between spatial resolution (improved by higher frequencies) and depth of investigation (reduced by the greater attenuation of high-frequency waves in lossy geological materials) that governs the design of near-surface acoustic tomography surveys applies equally to reservoir-scale crosswell seismic surveys; the iterative ray-path inversion methods used in surface and crosswell tomography share common mathematical foundations with computed X-ray tomography (CT), medical magnetic resonance imaging (MRI), and positron emission tomography (PET), reflecting the universality of the Radon transform as the mathematical tool for reconstructing a spatial distribution from its line integrals measured at multiple angles.