previous up next print clean
Next: IMAGING CONDITION AND LOCAL Up: OVERVIEW OF THE MIGRATION Previous: step 2

step 3

Backward time propagation of the recorded particle-displacement field and ${\bf w}$ and its time-derivative ${\bf \dot{w}}$, starting at the end time of the shot record and ending at time zero. These wavefields are introduced as time-dependent boundary conditions at z=0. Simultaneously with this backward propagation, this step also computes the lag zero of the correlation between the downgoing scalar field u(x,z) and the upcoming scalar field w(x,z). For each time step the following arrays are computed:

Finally, the reflection coefficient image is computed as
\begin{eqnarraystar}
\Psi(x,z) & = & {\int u(x,z,t) w(x,z,t) \, dt \over E(x,z)}...
 ...,z,t) w(x,z,t) \, dt \over E_{cut}}, \;\;\; 
\mbox{elsewhere,}\end{eqnarraystar}
and the local Snell parameter image as

\begin{displaymath}
\check{p}(x,z) = {\int \mid u(x,z,t) w(x,z,t) \mid \, \tilde{p}(x,z,t) 
\, dt \over \int \mid u(x,z,t) w(x,z,t) \mid \, dt},\end{displaymath}

where $\tilde{p}(x,z,t)$ is computed from ${\bf \hat{i}}$ and ${\bf \hat{r}}$,as described in the next section.


previous up next print clean
Next: IMAGING CONDITION AND LOCAL Up: OVERVIEW OF THE MIGRATION Previous: step 2
Stanford Exploration Project
12/18/1997