next up previous print clean
Next: Experiments Up: Shragge and Artman: Imaging Previous: Introduction


In this section, we outline the steps for preparing data from buried sources for imaging. Figure [*] illustrates the work flow. Central to each of these new applications is the careful attention to the geometry and characteristics of the source wavefield. To be able to utilize these alternative sources, in many cases the first stage is determining when and where an event has occurred. The primary energy mode (i.e., P or S-wave) and the direction of first motion must then be determined to enable characterization of the available scattered phases. After this information is ascertained, an appropriate source wavefield is constructed for input to the migration.

Figure 1
Flow chart illustration the steps to prepared data from buried sources for multi-mode migration. Solid line shows overall direction of processing flow. Dashed lines show dependancies.

After identifying source parameters, time zero and the total length of the record to be migrated is determined. Three-component receiver arrays are necessary to exploit the full range of wavefield scattering combinations in these experiments. With the introduction of vector displacements, a rotation of data components is defined that maps energy from the recording geometry [Z,N,E] to an alignment with source wave vector axes [P,SV,SH].

The Table below summarizes all of the information required to generate images from various combinations of scattering modes for a source with an initial P-wave polarization. In the first column, FS and BS represent forward- and backscattering, respectively. Forward-scattered wavefields are introduced in a manner akin to the exploding reflector model of seismic imaging. The forward-scattered source wavefield propagates through the model space in the same direction as the data wavefield. This introduces the only substantiative modification to the shot-profile algorithm. This is implemented by a change in the sign of the exponential in the SSR equation (column 2). Additionally, the interaction of the primary source with the free-surface gives rise to downward reflected P- and converted S-wavefields that are, in turn, back-scattered in a manner similar to conventional reflection experiments. The second field in the first column indicates the various mode transitions available for use in migration. These are abbreviated by their source and receiver phases.

The third column contains the source and receiver velocities required to produce the desired image. The final column identifies the receiver component in which the individual scattering components are expected. Notice that there are a number of different modes in the P and SV sections and, accordingly, cross-mode contamination will occur.

Scattering Mode Source Prop. Dir. S. Velocity R. Velocity Rec. Component
FS P-P - P P $\overline{P} $
FS P-S - P S $\overline{SV}$
BS P-P + P P $\overline{P} $
BS P-S + P S $\overline{SV}$
BS S-P + S P $\overline{P} $
BS S-S + S S $\overline{SV}$
BS S-S + S S $\overline{SH}$

The ability to realize these conditions is highly dependent on the nature of the candidate buried source. Specifically, imaging with local earthquakes and induced micro-fractures requires a prior inversion for a source's nucleation location and rupture time. Additional issues are the correct identification of dominant energy mode and polarization, and perhaps a deconvolution of source functions of potentially significant complexity. In applications to reservoir monitoring or the imaging of a major fault zone (e.g. San Andreas), a proliferance of 3-C receivers at the surface and incorporation of existing velocity model provides background information to locate and characterize micro-tremor sources. Down-hole geometries make realizing these conditions somewhat easier since the spatial and temporal location of the source are known.

next up previous print clean
Next: Experiments Up: Shragge and Artman: Imaging Previous: Introduction
Stanford Exploration Project