next up previous print clean
Next: Computing aspects Up: Guitton: Multiple attenuation Previous: What will I do?

Marine Data Results

This section presents results obtained using the HRT on a the Mobil AVO data set Lumley et al. (1995). These prestack data are heavily contaminated by free-surface and water-bottom multiples and thus constitute a challenging test-bed for true amplitude multiple attenuation techniques. The strategy for the multiple elimination is as follows:
1.
Create a velocity panel with the iterative Hyperbolic Radon Transform using the Huber solver.
2.
Define a corridor between multiples and primaries in the velocity space.
3.
Mute out the primaries in the velocity space.
4.
Model back the multiples in the data space.
5.
Subtract the multiples field from the input data.
6.
NMO and Stack.
This method is most suitable for preserving primaries energy. I compare l2, l1 and l1 with l1 regularization for the inversion in step 1. I decide to parameterize the problem using velocities and not slownesses.

 
next up previous print clean
Next: Computing aspects Up: Guitton: Multiple attenuation Previous: What will I do?
Stanford Exploration Project
4/27/2000