202 Seismic Imaging For conventional studies, the refraction method requires only the measurement of arrival times of first arrival waves (direct and refracted waves) to provide a geological model. Amplitudes are not commonly used in seismic refraction studies. In the case of an irregular interface, the analysis of the distortion of the head wave arrival allows the detection of wave interferences, which can be associated with the presence of fractures (second field example). In chapter 3, the task of imaging near-surface structures has been addressed with a few seismic tomography approaches. Several seismic field datasets are used to illustrate the ability of tomographic tools. The first field example concerns a transmission tomographic technique used for inverting first-arrival times, picked from a 3D surface seismic dataset, which was part of a geophysical survey conducted in a karstified dolostone region. Due to the limited azimuthal coverage of the surface data, the tomographic inversion produced a 3D model for the karst aquifer, with significance only in the upper epikarst region (up to 7 m deep). To overcome this image depth limitation, data collected with down-hole receivers were used simultaneously with those from surface geophones, which extended the image depth to the underneath low-permeability volume (up to 28 m deep). This 3D approach revealed a set of elongated furrows at the base of the epikarst and identified heterogeneities deep inside the low-permeability volume that may represent high-permeability preferred pathways for water inside the karst. In the second field example, the seismic data were collected by triaxial geophones in a cross-well experiment using a reflection tomography procedure, which enabled the imaging of a limestone reservoir at a depth of about 1,850 m. The raw field data were processed like conventional offset VSP data and the information present in the travel time of reflected S-waves was exploited for imaging between the boreholes. The imaging was achieved by time and depth transformations using a VSP-CDP stack, guided by a S-S ray tracing with a velocity model based on previous P and S VSP analysis. Although the reflection tomography did not provide an image with high frequency waveform, it successfully demonstrated the possibility of imaging between two wells from seismic data collected with conventional borehole multicomponent sensors and seismic waves generated by a low-energy source. Finally, through the application of a diffraction tomography approach, based on the Born approximation, we were able to produce quantitative elastic depth images from multi-component offset VSP datasets. One dataset was collected in the North Sea, and another from acoustic and multi-component borehole data collected at two different boreholes in the Paris basin. The target zone of the North Sea survey, which covered the reservoir area, is a rectangle extending from 50 m to 550 m east of the borehole, with depths from 3,400 m to 4,400 m. Geologists were able to interpret the estimations of three elastic parameters – P and S-wave velocities and density – that were produced by the diffraction tomography. Then, the top of the Brent reservoir could be delineated continuously away from the borehole, while it was also possible to interpret two faults. This chapter also includes a discussion on the assessment of the elastic depth
RkJQdWJsaXNoZXIy NjA3NzQ=