A pattern-based approach for multiple removal applied to a 3D Gulf of Mexico data set |
| |
| Authors: | Antoine Guitton |
| |
| Institution: | 3DGeo Development Inc., 4633 Old Ironsides Drive, Suite 401, Santa Clara, CA 95054, USA |
| |
| Abstract: | Surface‐related multiples are attenuated for one sail line and one streamer of a 3D data set (courtesy of Compagnie Générale de Géophysique). The survey was carried out in the Gulf of Mexico in the Green Canyon area where salt intrusions close to the water‐bottom are present. Because of the complexity of the subsurface, a wavefield method incorporating the full 3D volume of the data for multiple removal is necessary. This method comprises modelling of the multiples, where the data are used as a prediction operator, and a subtraction step, where the model of the multiples is adaptively removed from the data with matching filters. The accuracy of the multiple model depends on the source/receiver coverage at the surface. When this coverage is not dense enough, the multiple model contains errors that make successful subtraction more difficult. In these circumstances, one can either (1) improve the modelling step by interpolating the missing traces, (2) improve the subtraction step by designing methods that are less sensitive to modelling errors, or (3) both. For this data set, the second option is investigated by predicting the multiples in a 2D sense (as opposed to 3D) and performing the subtraction with a pattern‐based approach. Because some traces and shots are missing for the 2D prediction, the data are interpolated in the in‐line direction using a hyperbolic Radon transform with and without sparseness constraints. The interpolation with a sparseness constraint yields the best multiple model. For the subtraction, the pattern‐based technique is compared with a more standard, adaptive‐subtraction scheme. The pattern‐based approach is based on the estimation of 3D prediction‐error filters for the primaries and the multiples, followed by a least‐squares estimation of the primaries. Both methods are compared before and after prestack depth migration. These results suggest that, when the multiple model is not accurate, the pattern‐based method is more effective than adaptive subtraction at removing surface‐related multiples while preserving the primaries. |
| |
| Keywords: | |
|
|
