Q estimation based on logarithmic spectral areas with different high and low frequencies
An Yong♦1,2 and Wang Xin-You♦1,2,3
1. State Key Laboratory of Petroleum Resource and Prospecting, China University of Petroleum, 102249, Beijing, China.
2. CNPC Key Laboratory of Geophysical Exploration, China University of Petroleum, 102249, Beijing, China.
3. China Construction Northeast Design and Research Institute Co., Ltd. 110000, Shenyang, China.
摘要 地震波在粘弹性介质中传播时，地层对高频能量的吸收作用强于对低频能量的吸收，因此随着深度的增加和偏移距的增大，高频分量和低频分量的对数谱面积差值会逐渐增大。根据这一特征，本文提出一种基于高低频对数谱面积差法的Q 值估算方法（A method for Q estimation based on the logarithmic spectral areadifference of high and low frequencies，简称LSAD_LH），推导了品质因子Q 值与高低频对数谱面积之差的理论关系，并通过单层衰减介质模型证明该方法估算Q值的适用性。为进一步验证本文LSAD_LH 法对频带宽度选择和噪声的敏感度，利用包含有随机噪声的模拟地震记录，将LSAD_LH 法与谱比法（LSR）和对数谱面积差法（LSAD）两种方法比较，结果表明：LSAD_LH 法对频带宽度选择的依赖性不强；在抗噪性方面明显优于LSR 法且与LSAD 法具有相同的优势。因为LSAD 法受透射系数影响，为进一步凸显LSAD_LH 法的优势，本文将此方法应用于数值模拟的多层介质零偏VSP 资料和实际零偏VSP 资料，两种情况的应用进一步充分证明了LSAD_LH 法的适用性和相对于LSAD 法在高Q 值估算的准确性，并且克服LSAD 法受透射系数影响的缺点。
Abstract：
When a seismic wave propagates through subsurface viscoelastic media, the formation absorbs the high-frequency energy of the seismic wave more strongly than the lowfrequency energy. As the depth and the off set increase, the difference between the logarithmic spectral areas with high and low frequencies gradually increases. Based on this seismic wave characteristic, we have developed a novel Q-estimation method based on logarithmic spectral area difference of high and low frequencies (referred to as the LSAD_LH method). In this paper, we derive the theoretical relationship between the Q value and difference of logarithmic spectral areas with high and low frequencies and prove the applicability of the LSAD_LH method using a single-layer medium numerical model. To verify the sensitivity of the LSAD_LH method to bandwidth selection and noise, we compare the LSAD_LH method with two credible methods—the logarithmic spectral ratio (LSR) and logarithmic spectral area difference (LSAD) methods using a synthetic model containing the random noise. The results demonstrate that the LSAD_LH method is not highly dependent on bandwidth, and in terms of noise immunity, it is significantly better than the LSR method and has the same advantages as the LSAD method. To further highlight the advantages of the LSAD_LH method, we apply the LSAD_LH and LSAD methods to the vertical seismic profiling (VSP) numerical simulation of the multilayer media and the fi eld zero-off set VSP data. The application of the two cases proves the applicability of the LSAD_LH method and the accuracy of the high Q-value estimation relative to the LSAD method. Moreover, via the transmission coefficient, the LSAD_LH method overcomes the weakness of the LSAD method.
2021, Vol. 18 
