Abstract

CORRALES, Álvaro; ALMANZA, Ovidio  and  MONTES-VIDES, Luis. Removal of attenuation and dispersion in pre-stack gathers using inverse filter with quality factor estimated on instantaneous spectrum. Rev. acad. colomb. cienc. exact. fis. nat. [online]. 2014, vol.38, n.148, pp.278-286. ISSN 0370-3908.

Physical phenomena such as attenuation of high frequency components and velocity dispersion deteriorate seismic images hindering their interpretation and inversion. Removing the attenuation and dispersion of seismic records involves the use of inverse filters based on models that include the quality factor, Q, therefore, a robust method should be used to estimate Q and provide reliable values to improve the image of the subsurface value. The matching pursuit decomposition (MPD) generates a frequency spectrum for each sample of the trace (instantaneous), where reliable Q values are estimated without the inconveniences of spectral analysis with windows based on the Fourier and Gabor transforms. Since the phase component of the filter is inherently stable, compensation applies only to the amplitude. MATLAB algorithms that implement both the MPD and the stable inverse filter for removing the effects of attenuation and dispersion on pre and post-stack data were developed. The Q value estimated on synthetic seismograms and used with an inverse filter allowed correcting the distortion of the wavelet by retrieving the amplitude and phase spectra. The methodology was applied to a pre-stacked marine seismic line PSTM increasing the signal-noise ratio and significantly improving the temporal resolution records.

Keywords : Attenuation; dispersion; instantaneous spectrum; matching pursuit decomposition; inverse filter; quality factor.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )