Rock Physics of extensional Faults and their seismic imaging Properties
Abstract
Seismic waves will detect changes in density and velocity, and potentially additional information with respond to these parameters will be useful for the seismic imaging of the Earth. Knowledge of faults and fault systems is of great importance in reservoir characterization. This requires sub-surface mapping of faults, which can only be done with seismic data, except in limited well locations. The seismic resolution will then be the ultimate limitation for how small faults can be imaged, but due to noise and other effects, the small faults, from a seismic point of view, are hard to image.
The main emphasis of the work has been on building realistic faulted earth models, simulate seismic data collection and analyse the resulting synthetic seismic:
Fieldwork was carried out on several extensional faults, and hand specimens weresystematically collected and analysed in the laboratory to obtain P- and S-wave velocities, as well as density and mineral content. Analyses of these velocity results concluded that three different scenarios can be observed in close vicinity to faults:
• Low-Velocity-Zone: Increasing velocities as a function of distance from the fault plane.
• High-Velocity-Zone: Decreasing velocities as a function of distance from the fault plane.
• Constant velocity: Velocities appear to be independent of the distance from the fault plane.
Based on the above scenarios, three groups of earth models were designed: one group of models having displacements, one group having velocity changes and an additional group of models combining both the displacement and velocity changes as characteristics of a fault.
Synthetic seismic was simulated by using a Finite Difference scheme on the different earth models, and different migration algorithms have been tested. The resulting images were then systematized to achieve some general results:
• Both the displacement and the lateral velocity changes are shown to impact the seismic image.
• Discontinuity of reflectors in the footwall of a fault should be interpreted with care since they could be caused by lateral velocity changes.
• The spatial position of a non-imaged fault plane should be based on terminations of reflectors in the hangingwall and not in the footwall.
• Bending close to a fault plane should be interpreted with care, since such phenomena could be artificially caused by the migration algorithm.
The general observed effects of a fault having an area of lateral velocity changes are useful for anyone interpreting faults. More specific effects in the seismic image can be utilized if a direct comparison is made between migrated images which used the same algorithm.
Finally, a seismic data set from the Gullfaks Field in the North Sea was chosen in order to test if the results obtained from the synthetic data could be applied on real seismic data. More specifically, one fault at the Gullfaks Field was chosen to test if lateral velocity variations in the nearby area of a fault could be used to improve the seismic imaging of the fault. The result shows that a weak improvement of the imaging of the fault plane can be observed in the middle area of the fault.