The Time Domain Inverse Q Filtering Method Based On Effective Q Factor

Because anelastic absorption in stratum has a time-varying filtering effect on seismic wave, the energy of seismic wave attenuates and the velocity of seismic wave disperses with the increase in travel time, which can lead that continuation of reflectation event becomes poor and resolution of deep layer becomes significantly degraded. The main objective of seismic data processing is to obtain high resolution seismic profile. We must implement amplitude compensation and phase correction on seismic data about anelastic absorption and velocity dispersion. Inverse Q filtering is a commonly used compensation method. As the view of seismic wave propagation, inverse Q filtering can be viewed as reverse direction propagation of seismic wave, which extends the data recorded on the surface to the reverse direction and removes the layered Q filtering effect generated by earth media absorption.Layered Q inverse Q filtering method based on wave field continuation theory has been already presented for severeal years, however, using seismic data to construct layered Q model has many difficulties and the precision of results is bad. In order to overcome the difficulties of modeling layered Q value and enhance the efficiency of inverse Q filtering algorithm, an inverse Q filtering method based on the effective quality factor in time domain is presented. The weighted least-squares optimization strategy is introduced into designing the short operator of inverse Q compensation in time domain for calculating the optimum short operator, which can change large amount of multiplication operation in the frequency domain to less convolution operation in time domain. We adopt the table-driven scheme to split short operator construction from operation of inverse Q compensation for raising the efficiency of calculation. A new method of stabilization controlling is proposed, it can ensure the stability of arithmetic and satisfy the demand of precision for designing short operators.We present a new method of determining effective Q, which combines the generalized S transform with scanning analysis technique. Inverse Q filtering is adopted to compensating for seismic stacked profile by using a series of effective Q factors. We analyze a stacked trace chosen from the stacked profile by using generalize S transform to gain the time-frequency spectrum and peak frequency of Fourier spectrum. Time-frequency spectrum of a shallow strong reflection may be selected as a reference. It is compared with the time-frequency spectrum of target reflection in different depth. A standard is established for choosing accurately effective Q, which is that the spectrum width and peak frequency of Fourier spectrum is consistent with reference spectrum.The time domain inverse Q filtering method is applied to theoretical data and actual seismic data. The results show that time domain inverse Q filtering method can achieve the same compensation effect as the arithmetic in frequency domain and ensure stability and higher efficiency. It is fully enough to prove correctness and practicality of method.