Multiple Identification Based On Frequency Differences

Generally speaking, the main frequency of seismic signal will become lower as the travel time increasing because of the earth's absorption and attenuation. However, there can be significant differences in the changes of frequency spectrum between the reverberation and primary waves. Based upon a simplified marine seismic model with different medium quality factors along the different wave trajectories, this thesis proves that the main frequency varying rate of reverberation can be much lower than that of the primary wave. This concludes that on the middle and deep seismic record, the main frequency of reverberation can be much higher than that of primary wave within a same travel time window. The conclusion is further verified by different numerical results modelled using viscoacoustic wave theory. These numerical results demonstrate that the main frequency of multiple can be always higher than that of primary wave within a same travel time window.The attenuation of multiples is one of the difficult problems in the marine seismic data processing. The criterion to judge the multiple suppression quality is that the primary signal should be highlighted with the multiples being attenuated. Whereas eliminating multiples to a good result they must be pre-identified. Traditional method to identify a multiple is mainly according to the travel time in the seismic record (this is referred to Time Identification Mode). Based upon the theoretical and numerical analysis, the thesis presents a new approach, the Time-Frequency Identification Mode, in which a wave is recognized to be a multiple not only by the travel time but also the main frequency range of the signal. Real seismic data processing examples demonstrate that using the Time-Frequency Identification Mode can suppress the multiple effectively and meanwhile protect the basic characters of the primary wave. Whereas there is a quite possibility of damaging the primary waves if only the Time Identification Mode is used in the process of multiple attenuation.