previous up next print clean
Next: WEIGHTED LEAST-SQUARES Up: GENERAL FORMULATION OF THE Previous: GENERAL FORMULATION OF THE

Testing the method on a synthetic dataset

To illustrate the method and compare the results from the different schemes described here, we have generated a simple synthetic dataset using an elastic modeling program. Figure [*]-a shows the reverberation-free data obtained with the elimination of the surface (air-water) boundary. The pure P wave reflections are the events at approximately 0.4, 1.6, and 3.1 seconds. Some converted PSPP and PSSP reflections are also visible at the far offsets between 2.0 and 3.0 seconds, and at about 4.5 seconds. The more realistic dataset (with the surface boundary present), which has been used as input for the several multiples suppression schemes, is shown in Figure [*]-b. Figure [*]-a shows the semblance spectrum of the data overlaid with the manually designed window in which reverberations are dominant. The result of applying the basic scheme described by equation (4) to the data in [*]-b is shown in Figure [*]-b.

 
syntdata
syntdata
Figure 1
(a) Synthetic data, free of surface-related reverberation. (b) Synthetic data with surface-related reverberations.
view burn build edit restore

 
semrm0
semrm0
Figure 2
(a) Semblance panel of the data of Figure [*]-b, overlaid with the manually designed window function. (b) Result of applying the scheme defined by equation (4) to the same data.
view burn build edit restore


previous up next print clean
Next: WEIGHTED LEAST-SQUARES Up: GENERAL FORMULATION OF THE Previous: GENERAL FORMULATION OF THE
Stanford Exploration Project
12/18/1997