Details

Autor(en) / Beteiligte

• Gholamy, Shabanali
• Javaherian, Abdolrahim
• Ghods, Abdolreza

Titel

Automatic detection of interfering seismic wavelets using fractal methods

Ist Teil von

• Journal of geophysics and engineering, 2008-09, Vol.5 (3), p.338-347

Ort / Verlag

IOP Publishing

Erscheinungsjahr

2008

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Interference of reflections from upper and lower interfaces of geological units is a common source of interpretational ambiguity. This paper utilizes the fractal dimension of instantaneous seismic attributes to study this interfering phenomenon. Using a 1D synthetic seismic trace, we examine the divider, Hurst, spectral and variance fractal methods. Since each method utilizes a moving window technique as part of its analysis, a series of optimum window lengths needs to be determined. The four methods are applied to the instantaneous amplitude, phase and frequency attributes of a series of 1D synthetic seismic traces. To study wavelet interference on 2D seismic sections, we generate four 2D synthetic seismic sections by convolving a zero phase and a minimum phase wavelet with two different wedge models. For the first model, the reflection coefficient of the upper wedge interface is positive and the lower interface negative, whereas for the second model, the polarities are positive. Our results suggest that fractal analysis of the instantaneous phase, using the Hurst method, can separate two opposite polarity wavelets within a distance of l/16 or greater and of l/4 or greater when they have the same polarity. Our results are not sensitive to random noise when the signal-to-noise-ratio is greater than 10.