Search

The problem of localizing small (relative to wavelength) scatterers by diffractions to enhance their use in identifying small-scale details in a seismic image is extremely important in shallow exploration to identify interesting features such as fractures, caves, and faults. The conventional approach based on seismic reflection is limited in resolution by the Rayleigh criterion. In certain acquisition geometries, such as crosswell surveys aimed at obtaining high-resolution signals, the availability of suitable data sets for effective migration depends on the spatial extent of the available source and receiver data intervals. With the aim of overcoming the resolution limits of seismic reflection, we study the detectability, response, and location of meter- and possibly submeter-dimension carbonate concretions (septaria) in the Boom Clay Formation (potential host rocks for radioactive waste disposal) by diffraction analysis of high-frequency signals. We investigate diffraction wavefields by signal separation, focusing, and high-resolution coherency analysis using the multiple signal classification method and semblance. The investigation is performed for two different surveys in Belgium, a shallow and high-resolution reverse vertical seismic profile (RVSP) and a near-offset crosswell application at Kruibeke and ON-MOL-2 sites, respectively. The data analysis is supported by synthetic wavefield modeling. The multioffset RVSP provides the appropriate geometry to observe and investigate the septaria diffractions from depth and the surface. The crosswell approach, calibrated using synthetic data in the analysis of wavefield patterns in 2D, indicates promising imaging results with field data of a selected diffraction zone in the interwell area.

For four decades, the use of the drill-bit seismic source has been evaluated extensively and investigated for seismic-while-drilling (SWD) purposes. After an initial diffusion phase, however, the routine application on drill-bit reverse vertical seismic profiling (RVSP) has declined because of issues related to variability in the signal-to-noise ratio of drill-bit signals obtained with different drilling conditions. Consequently, additional efforts have been dedicated to improving the recording of reference signals. Recent technology developments enable direct ground-force pilot measurements at downhole positions close to the bit, which enable reliable estimations of the far-field drill-bit-radiated signature. The knowledge of the ground force below the source in the near-field is a key aspect not only for the exploitation of the drill-bit source but also for conventional seismic vibrator sources. The SWD deconvolved results obtained by the ground force are different from those obtained by recording only motion vibration, such as acceleration. The ground force and particle velocity are linked by a complex impedance in the near-field, and together they provide dual measurements. In addition, the ground force provides signals with effective coupling conditions at the bit-rock interface in the presence of the drill-bit displacement action with penetration during rock fracturation. The analysis of the impedance at the bit source includes the drillstring reflections, which are observed in the reference drillstring pilot signal and are simulated by using an appropriate mechanical model. The processing of real seismic signals with different approaches by deconvolution and correlation using ground-force and motion signals produces improvements in the extraction of the source signal, with effective removal of the undesired pilot wavefields, and with cleaner estimations of the drill-bit seismic wavefields in a crosswell survey.

Distributed acoustic sensing (DAS) using fiber optic cables is an emerging seismic acquisition technology for the oil and gas industry, geothermal resource exploration, and underground fluid-storage monitoring. This technology offers the advantage of improving seismic acquisition by enabling massive arrays for monitoring of seismic wavefields at reduced cost with respect to conventional methods. In general, it is accepted that this method provides acoustic signals comparable with conventional seismic data, however, without the multicomponent directional information typical of geophones. We have developed a modified data extraction method and found that, as a result of the dense spatial distribution of recording points along the optic cable, DAS can provide two linked wavefield components in the axial direction, even when using a single 1D cable line. These signal pairs consist of dual components that are related to native strain rate (or strain) and particle acceleration (or velocity) fields at a given recording location. These dual signals are easily usable for wavefield separation purposes simply performing a trace-by-trace combination by appropriate scaling coefficient. The analysis performed with borehole data from linear and helically wound cables demonstrates the effectiveness of polarity recovery and dual-wavefield separation. We show real examples in which the data can be combined to provide separation of up- and downgoing wavefields. The ratio of the dual components provides information on local slowness properties in the formation.

The virtual-reflector (VR) method creates new seismic signals by processing seismic traces that have been produced by impulsive or transient sources. Under proper recording-coverage conditions, this technique allows a seismogram to be obtained as if there were an ideal reflector at the position of the receivers (or sources). Only the reflected signals from this reflector are synthesized. The algorithm is independent of the medium-velocity model and is based on convolution of the recorded traces and on subsequent integration of the crossconvolved signals in the receiver (or source) space. We use the VR method in combination with seismic interferometry (SI) by crosscorrelation to compose corresponding virtual-reflection events in seismic exploration. For that purpose, we use weighted-summation and data-crossfiltering approaches. In applying these combination methods, we assume common travel paths in the virtual signals, taking into account that VR and SI by crosscorrelation imply different stationary-phase conditions. We present applications in which we combine the SI-by-crosscorrelation and VR signals to (1) suppress unwanted effects, such as marine water-layer reflections in synthetic ocean-bottom-cable data, and (2) obtain virtual two-way traveltime seismograms with real borehole data from walkaway vertical seismic profiling (VSP). Analysis shows that time gating and selection of reflection events are critical steps in processing water-layer multiples.