Seismic Absorption Parameter Inversion And Absorption Compensation Based On Self-Adaptive Signal Identification

Amplitude attenuation and velocity dispersion of seismic waves in viscoelastic media seriously reduce seismic resolution and detection depth.Therefore,the study of seismic attenuation is very important for high-resolution seismic exploration.Q factor inversion is the main tool to establish the absorption model,which is the basic dataset for implementing absorption compensation.Furthermore,seimsic absorption compensation is the fundamental way to improve seismic resolution.Spectral ratio method(SR)and centroid frequency shift method(CFS)are two widely used Q-factor inversion methods in seismic exploration.They mainly rely on a single attribute(e.g.logarithmic spectral ratio slope or centroid frequency)of the seismic spectral to estimate Q factor.Therefore,the inversion results are easily affected by factors such as seismic noise and wavefield interference.To relieve this problem,we propose a novel Q-factor inversion method using the source spectral consistency(SSC).In this method,seismic signals are backwards propagated and compensated to their source position along their rays,then,the Q factors are estimated based on the spectral consistency of the reconstructed source signals.Compared with the Q-factor estimation methods based on centroid frequency,the source signature can be unknown and it is unnecessary to extract a specified attribute of the signal spectral.Therefore,the proposed method has a great potential to improve the Q-factor inversion accuracy.The experimental results show that the Q-factor inversion method based on the source spectral consistency has better anti-noise property and higher inversion accuracy,and can provide a higher-precision Q model for subsequent absorption compensation.Based on the estimated Q model,we can use inverse-Q filtering to eliminate the seismic absorption effects and improve the seismic resolution.However,the traditional inverse-Q filtering can not identify seismic signal in the algorithm,which will result in the amplification of seismic noise.To alleviate this issue,we propose a new multichannel absorption compensation method based on self-adaptive signal identification.The core of this apporach is to introduce a signal characterization operator,which can describe the geometric pattern and amplitude variation of seismic reflections,into the multichannel absorption compensation inversion system.In the inversion process,seismic signal and noise are automatically identified and selectively compensated.Thus,the noise effects on absorption compensation are systematically removed and the accuracy of absorption compensation results is improved.Considering that the reflection structure characteristics of seismic signals can be described in both the time-space domain and the frequency-space domain,therefore,we implement the signal-identification multichannel absorption compensation in both the time domain and the frequency domain.Then,we compare the differences and connections between the two apporaches.The synthetic data tests and field data examples show that the proposed methods can not only improve the resolution of seismic data,but also protect the spatial structure characteristics and has strong stability and certain practicability.Since the absorption effects occur in the seismic wave propagation,it is physically reasonable to perform absorption compensation in the migration.Seismic absorption compensation in the wavefield extrapolation will inevitably cause the amplification of high-frequency noise,which redues the migration stability and imaging accuracy.To solve this problem,we use the Kolsky-Futterman model to derive a frequency-domain viscoacoustic wave equation,and then implement a stabilized Q-compensated reverse time migration in the frequency domain.In the algorithm,we first derive a frequency-domain viscoacoustic wave equation,which has decoupled amplitude attenuation and phase dispersion effects.Then,we use the dispersion-only and viscoacoustic wavefields to construct a stabilized attenuation compensation operator for the seismic wavefield extrapolation.Finally,the frequency-domain cross-correlation imaging condition is exploited to obtain the compensated image.The synthetic experiments show that the proposed method can not only compensate for the absorption effects,but also suppress the amplification of high-frequency noise and improve the migration stability and imaging quality.