Hydrocarbon detection using adaptively selected spectrum attenuation

Hydrocarbon reservoir usually shows the characteristics of low frequency amplifying and high frequency attenuating, which has been used as hydrocarbon indicator. However, reflection interference may also cause these phenomenons. In this paper, we propose an adaptively selected spectrum attenuation method to reduce the impact of the reflection interference in hydrocarbon detection. We first evaluate spectrum attenuation of seismic data by calculating the ratio of high frequency to low frequency components for each trace in the time-frequency domain. Notice that high frequency components are not really attenuated by interference as they appear, they can be recovered later, so we use an edge-preserving smoothing (EPS) method to smooth the equivalent local peak frequencies (ELPFs) located at the local peaks of seismic envelope to measure the attenuation trend of a seismic trace, and choose the spectrum attenuation whose high frequency components are really attenuated to locate the hydrocarbon. A real seismic data example demonstrates that the spectrum attenuation calculated by the ratio can avoid the impact of strong reflection, and the selected spectrum attenuation can reduce the effect of interference, which makes it more reliable in locating hydrocarbons than use the spectrum attenuation directly.