--> ABSTRACT: Wavelet Dispersion and Bright-Spot Detection, by Peter C. Luh; #91022 (1989)
[First Hit]

Datapages, Inc.Print this page

Wavelet Dispersion and Bright-Spot Detection

Peter C. Luh

Since Ostrander in 1984 showed that the variations in reflectivity as a function of Previous HitoffsetNext Hit can be used for bright-spot validation, it has been known that these variations are often sensitive to errors in acquisition and processing of prestack seismic data. This presentation shows that even with perfect measurements, the bright-spot signal will be altered because the wavelet disperses as it propagates through an attenuating overburden. This in turn affects the slope term of the amplitude vs. Previous HitoffsetNext Hit (AVO) variation. For any single reflector, the dispersion of its propagated signal as a function of Previous HitoffsetNext Hit can be decomposed into a product of the vertical and horizontal dispersions. The vertical dispersion is the dispersion that a zero-Previous HitoffsetNext Hit arrival suffers through its propag tion, and the horizontal dispersion is the additional dispersion that a nonzero-Previous HitoffsetTop arrival must suffer further because of the residual normal moveout time. Numerical examples using a frequency-independent attenuation law show that even for a relatively high-Q or low-loss overburden, the horizontal dispersion alone can distort the AVO signal. This distortion cannot be taken care of by velocity analysis. A preferred method to overcome the dispersion effect would be to apply Q compensation on prestack seismic data before movement.

AAPG Search and Discovery Article #91022©1989 AAPG Annual Convention, April 23-26, 1989, San Antonio, Texas.