Multiple attenuation in the parabolic tau-p domain using wavefront characteristics of multiple generating primaries

Citation
E. Landa et al., Multiple attenuation in the parabolic tau-p domain using wavefront characteristics of multiple generating primaries, GEOPHYSICS, 64(6), 1999, pp. 1806-1815
Citations number
20
Categorie Soggetti
Earth Sciences
Journal title
GEOPHYSICS
ISSN journal
00168033 → ACNP
Volume
64
Issue
6
Year of publication
1999
Pages
1806 - 1815
Database
ISI
SICI code
0016-8033(199911/12)64:6<1806:MAITPT>2.0.ZU;2-L
Abstract
The problem of multiple attenuation has been solved only partially. One of the most common methods of attenuating multiples is an approach based on th e Radon transform. It is commonly accepted that the parabolic Radon transfo rm method is only able to attenuate multiples with significant moveouts. We propose a new 2-D method for attenuation of both surface-related and inter bed multiples in the parabolic tau-rho domain. The method is based on the p rediction of a multiple model from the wavefront characteristics of the pri mary events. Multiple prediction comprises the following steps: 1) For a given multiple code, the angles of emergence and the radii of wave front curvatures are estimated for primary reflections for each receiver in the common-shotpoint gather. 2) The intermediate points which compose a specified multiple event are det ermined for each shot-receiver pair. 3) Traveltimes of the multiples are calculated. Wavefields within time windows around the predicted traveltime curves may b e considered as multiple model traces which we use for multiple attenuation process. Using the predicted multiple traveltimes, we can define the area in the tau-rho domain which contains the main energy of the multiple event. Resolution improvement of the parabolic Radon operator can be achieved thr ough a simple multiplication of each sample in the tau-rho space by a nonli near semblance function. In this work. we follow the idea of defining the multiple reject areas auto matically by comparing the energy of the multiple model and the original in put data in the tau-rho space. We illustrate the usefulness of this algorit hm for the attenuation of multiples on both synthetic and real data.