APPLIED GEOPHYSICS,Vo1.8,No.4(December 201 1),P.344-354 7 Figures DOI:IO.1007/s1l770—011-0296一y An analysis of seismic scattering attenuation in a random elastic medium★ Liu Jiong1,2,3,Wei Xiu-Cheng’ ,,Ji Yu-Xin’ ,,Chen Tian-Sheng’, ,Liu Chun—Yuan’, ,Zhang ChunITao’ ',and Dai Ming。Gang Abstract:Seismic attenuation has been an interesting topic of research,for it reflects the inherent media characteristics in which seismic waves propagate.There are many factors that cause seismic wave attenuation,such as geometry attenuation caused by energy dissipating during propagation,friction attenuation by relative sliding among rock grams, and scattering attenuation by rock heterogeneity.In this paper we study P—wave scattering attenuation in a random elastic medium by numerical simulations from a statistical point of view.A random elastic medium model is built based on general stochastic process theory. Then a staggered—grid pseudo—spectral method is used to simulate wave propagation. Scattering attenuation is estimated by the spectral ratio method based on virtual detector records.Random elastic media numerical scatter results with various heterogeneity levels show that the higher heterogeneous levels cause greater scattering attenuanon.When the scatter sizes are smaller than a wave length.the larger scatters give a greater attenuation. Finally.we propose a method to evaluate fluid—flow attenuation in porous media.The fluid— low attenuation is derived from totalf attenuation and scattering attenuatmn in random porous media and the attenuation is estimated quantitatively.Results show that in the rea1 seismic frequency range when the heterogeneous scale is about l 0 meters(Iess than one wave length).scattering attenuation is larger than fluid—lfow attenuation in random porous media and scattering attenuation is the main factor of seismic attenuation in real heterogeneous porous media. Keywords:scattering attenuation,random elastic medium,the spectral ratio method Introduction seismic wave attenuation refers to the energy Ioss during the propagation in media.Since it can reflect seismic wave attenuation,such as geometry attenuation caused by energy dissipation during wave propagation, friction attenuation by relative sliding among rock grains,scattering attenuation by rock heterogeneity, fluid—flow attenuation.and etc.Because of the complexity in the seismic attenuation mechanism,the the inherent media characteristics in which the seismic wave propagates,seismic attenuation has been a main concern for many researches,such as exploration understanding and application of seistoic attenuation iS still at early stage. scismo1ogY,rock mechanic s,non.destructive inspection and etc.There are many factors causing Chemov f 1 960)researched seismic wave attenuation by scalar wave propagation theory in random media, Manuscript received by the Editor January 7,201 1;revised manuscript received October 3,20 1 1. This research is supported by the State Key Program ofNational Natural Science of China fGrant No.4083990 1) 1.Sinopec Key Laboratory of Multi-Component Seismic Technology,Beijing 1 0083,China. 2.Exploration&Production Research Institute,SINOPEC,Beijing 10083,China. 3.Resource and Communication College,China University ofPetroleum(Beijing),Beijing 102249,China. 344 Liu et a1. which ma ̄ed the beginning of scattering research.Aki r 1 969)simulated the formation of seismic coda based on single scalar wave scattering theory and concluded Principle and method that the coda produced by random heterogeneity is the main reason why the earth sustains 1ong—time vibration. Random medium lheOr\, Previous researches on seismic wave propagation Aki f 1 973)and Berteusen ct a1.f l 975)used scalar wave scattering theory to interpret the P—wave phase and amplitude fluctuations observed in large aperture seismic focused on homogeneous or layered homogeneous media. However,for a large number of irregularly—distributed heterOgeneOus media,a random medium model is the arrays far from the source.Wu r1 982)developed a multi—scattering scalar wave approximation for forward propagation to interpret S—wave frequency character and coda attenuation. Thereafter have followed many complex elastic wave researches.such as studies Of 11 elastic wave scattering theory in solid media with small horizon heterogeneity, and many meaningful conclusions were made(Sato. 1982a and 1982b;Wu and Aki 1985a,1985b,1988;Wu, 1 989).Wu r 1 994)developed the elastic wave scattering theory in media with large horizon heterOgeneitV.Li and Hudson(1 9961 developed elastic wave multi. scattering theory in media with larger heterogeneity and considering the scattering attenuation. However,these seismic scattering theories for seismic wave propagation in heterOgeneOus media were all deftved with approximation assumptions.In reality,these assumptions may not be satisfied,so the applications are limited.In this Paper,beyond those limits.numerical modeling like Frankel and Clayton f 1 986)is used to research scattering attenuation without the theoretical assumption limitations. We focus on scattering attenuation in random elastic media:First,wave propagation is simulated and then the spectra1 ratio method in the frequency domain is adopted to estimate the P—wave scattering attenuation in the media.Since media properties differ in different directions.the mean wave attenuation value in a few directions is used to scale attenuation in the whole random elastic medium.The stochastic process determines that different realizations lead to different scattering attenuation.So several random elastic models with the same stochastic character are constructed to analyze the attenuation.Liu et a1.(20 1 O1 studied seismic attenuation in random porous media and pointed out that seismic attenuation is caused by two factors of fluid flow and elastic scattering but the two attenuation mechanisms were not separated.In this paper,we propose a new method to scale the fluid. flow attenuation in random porous media.The tota1 attenuation and scattering attenuation are computed quantitatively,from which the fluid—flow attenuation is derived quantitatively. better choice to describe the actual situation. According to stochastic process theory,an arbitrary variable g distributed in two—dimensional space can be expressed as the sum of the mean value and perturbation rXi and Yao。2002) g(x,z)=go(1+r(x,z)), (1) where go is the mean ofg in space(a constant)and), ,Z) is the perturbation relative to the mean value at the point ,z).For mathematical convenience and assuming that the random spatial perturbation be a stationary stochastic process with mean equal to zero,its expression is <r(x, )>=0 (2) In addition to mean,two other parameters,variance and auto—correlation function,are used to describe stationary stochastic processes.They are expressed as: o- =<y2 ,z)>, (3) ( ,z)=< ,zOr( +x,zI+z)/:, (4) where<‘>denotes the mean operator. According to stochastic process theory,the), ,z) power spectrum is the Fourier transform of its auto. correlation function ,z).So we can use the auto. correlation and spectral expansion method to construct the random spatial distribution of r(x,z)(Ikelle et a1., 1 993).There are a variety of auto—correlation functions that can be selected when cOnstructing a random model,such as Gaussian,exponential,and Von Karman function.The exponential auto—correlation function is adopted in this PaDer and the expression iS ( ,z)=e (5) where a and b are the auto—correlation lengths in the x and z directions,respectively.In a random medium, greater a and b indicate greater heterogeneiyt. After perturbation r(x,z)is computed by the spectral 345 Seismic scattering attenuation expansion method,the random variable g , in two— In a random elastic medium.medium properties in dimensional space can be determined by equation(1). adiacent regions may have signiicantf differences.which easily results in non.convergence in the iterations during Random elastic mode In an elastic medium,seismic wave equations include geometry equations for strain,kinematic equations,and constitutive equations.The kinematic equation expressed wave—field simulation.On the premise of guaranteeing the simulation accuracy,the staggered—grid pseudo— spectral method can identify strong heterogeneity in = = space(Ozdenvar and McMechan 1 996). by Cartesian tensors is Spectral ratio method 、 ●/\ ●/\、●oij — puf ■■一 ox where“is the particle displacement 口is the stress and Fisthebodyforce. For an isotropic medium,we replace the geometry equations into the constitutive equations and get = 薏 + (考+善3 j(7 where and are Lam6 coefficients and 6 lj is the Kronecker tensor,which is 1 when i=j and 0 when i4:j. When an elastic wave propagates in the x—z plane,we put the corresponding conditions into equations(6)and (7)to derive the controlling equations: J I :/:x -8d仃 .oo  ̄斗+ 8o--= (8) and In this paper.rock density P and Lam6 coe所cients and are all assumed to be independent of each other and are randomly distributed jn space.For mathematica1 cOnvenience.we assume that all the random variables have the same zero average perturbation and the same variance and exponential auto.correlation function.In stochastic processes,the mean square deviation(MSD) o-can reflect the degree of random variables deviated from the mean value.For greater .the heterogeneity is higher.So we can get random elastic properties at different heterogeneous 1evels by changing o-. 346 Generally,seismic wave amplitude can be expressed + 一 + 一 2 2 as(Toksoz et a1.,1 979) A(J’、=G(x)e e ‘ f{;丝 where A is the amplitude in the ffequencv domain. G 1 is the geometrical diffusing factor independent 0f frequency(Aki and Richards,2002;Gusmeroli et a1.,20 1 0),x is the wave propagation distance,/’is the rfequency,o[(n is the attenuation coefifcient related wim rfequency,and k is wave number. The spectra1 ratio method can be used to estimate seismic wave attenuation based on thc records at different points(Tonn,1 99 1 1.In this method,first we extract two points, 】and x2,from the same wave propagation path.Then we substitute Xl and x2 into equation(10)to get A1 and A,.We take the natural logarithm of the T.csult of A divided by A,.The final expression is f( .1l<-ione et a1..2003): cX2-XI n导 In addition to attenuation COCfficient.the inverse quality factor 1/Q is often used to scale sei smic attenuation.The inverse quality factor is defined as l l△ ÷一. D 2丌E where E is the peak energy in a period and AE is the lost energy at the corresponding time.If the attenuation coefifcient is much less than 1 in a wavelength,the inverse quality can be approximately expressed as: Q 兀} l 一≈——. (13) where V is the plane wave phase velocity.Substituting equation(13)into equation(1 1),we get dl-一 0I zrf …n导G 一Liu et a1. with controlling equations(1 7)to(22)are performed and the scattering attenuation results are shown in Table 9.The fluid-lfow attenuation inverse quality factors are shown Table 10. Table 8 Total inverse quality factors in the random porous medium From Tables 8 to l 0.we see that the total attenuation levels and with diferent scatterer size and they are used to estimate the inverse quality factor by the spectral ratio mean value in the five tests is 0.07244,the scattering attenuation mean value is 0.062 1 4.and the fluid flOW attenuation mean value iS 0.01032.Scattering attenuation method.In addition,a method that estimated fluid—flow attenuation is proposed and used to compute fluid—flow attenuation. is stronger than lfuid—flOW attenuation and their ratio is approximately 6:1.This shows in our simulations.when Numerical results show that.in the random elastic model,seismic waves will experience attenuation due to scattering during propagation.Higher heterogeneous the scale of heterogeneity is 1 0 m.attenuation caused by scattering is stronger than that of lfuid-low fin seismic rfequency and scattering attenuation is the major part of totaI attenuation. levels results in greater attenuation.When scatter sizes are less than a wavelength,larger scatters yield stronger attenuation.This conclusion is consistent with scattering theory.Numerical results in random porous media show COnCluSiOnS Based on elastic wave and stochastic process theories, we design a random model with heterogeneous elastic structure and adopt the staggered・-grid pseudo—-spectral method to simulate seismic wave propagation in the mode1.Virtual detector displacement records of the wave—fields are generated at different heterogeneous that when source frequency is low and heterogeneity size is on the scale of 1 0 m.fluid—flOW attenuation is less than scattering attenuation at seismic frequency and their ratio is approximately l:6. In reality all heterogeneity scales exist in rock.The scales range from being much less than a wavelength to being much greater than a wavelength.So scattering attenuation iS one of the major sources that causes strong seismic wave aRenuation in real rock strata. 353 Seismic scattering attenuation ACknOWIedgementS We are grateful to Dr.Ba Jing for his help during the course of this work.We also want to thank the anonymous reviewers for their thoughtful comments and valuable (in Chinese):Oil Geophysical Prospecting,43,723—727. Mavko,G.,Mukerji,T.,and Dvorkin,J.,l 998,The rock physics handbook:Tools for seismic analysis in porous media:Cambridge University Press,London. 0zdenvar,T.,and McMechan,G.A.,1 996,Causes and reduction of numerica1 artifacts in pseudo.spectral suggestion,which greatly improved the manuscript. References Aki K.,l 969,Analysis of the seismic coda of logical earthquakes as scattered waves:Journal of Geophysical Research.74.6l5—631. Aki,K.,1 973,Scattering of P waves under the Montana LASA:Jouma1 ofGeophysical Research.78 1334—1346. Aki,K.,and Richards,P.G.,2002,Quantitative seismology. Second edition:University Science Books,Sausalito,CA. Berteussen,K.A.,Christoffersson,A.E.S.,Husebye, E.S.,and Dahle,A.,1 975,Wave scattering theory in analysis of P wave anomalies at NORSAR and LASA: Geophysical Journa1 Royal Astronomical Society 42 403—417. Carcione,J.M.,Helle,H.B.,and Pham,N.H.,2003, White’s model for wave propagation in partially saturated rocks:Comparison with poroelastic numerica1 experiments:Geophysics.68(4).1389一I398. Chemov,L_A.,1 960,Wave propagation in a random medium:McGraw—Hil1 Press New York. Frankel,A.,and Clayton,R.W. 1 986,Finite—diffeFence simulations of seismic scattering implications for the propagation of short—period seismic waves in the crust and models of crusta1 heterOgeneitv:Journa1 of Geophysics Research,91.6465 6489. Gusmeroli,A.,Clark,R.G.,et a1.,20l 0,Seismic wave attenuation in the uppermost glacier ice of Storglaciaren. Sweden:Journal of Glaciology 56.249—256. Ikelle,L_T.,Yung,S.K.,and Daube,F. 1993 2一D random media with ellipsoidal autocorrelation function: Geophysics,58.1359~1372. Li,X.,and Hudson,JJ A.,1 996,Multiple scattering of elastic waves from a continuous and heterogeneous region:Geophysical Journa1 Internationa1.126.845—862. Liu,J.,Ba,J.,Ma,J.W.,et al,2010,An analysis of seismic attenuation in random porous media:Science in China Series G.53 628 637. Liu J.,Ma J.W.,Yang H.Z.,et.al,2008,Staggered g d pseudo—spectrum simulation of rfacture and cave reservoir 354 wavefield extrapolation:Geophys J Int.126 8l9 828. Sato,H.,1982a,Amplitude attenuation of impulsive waves in random media based on travel time corrected mean wave formalism:Journa1 of Acoustic Society of America. 71,559 564. Sato,H.,1982b,Attenuation of S waves in the lithosphere due to scattering by it’s random velocity structure: Journal of Geophysics Research.87.7779—7785. Toksoz,M.N.,Johnston,D.H.,and Timur,A.,1979, Attenuation of seismic waves in dry and saturated rocks: I.Laboratory measurements:Geophysics.44.68 l 690. Tonn,R..1 99 1.The determination of seismic quality factor O from VSP data:A comparison of different computational methods:Geophys.Prosp..39.1—27. Wu,R.S..1 982.Attenuation of short period seismic waves due to scattering:Geophysics Research Letters 9.9—1 2. Wu,R.S.,and Aki,K.,1 985a,Scattering characteristics of elastic waves by and elastic heterogeneity:Geophysics, 50 582——595. Wu,R.S.,and Aki,K.,1 985b,Elastic wave scattering by a random medium and the small—scale inhomogeneities in the lithosphere:Journal of Geophysics Research,90 10261一l0273. Wu,R.S.,and Aki,K.,1 988,Multiple scattering and energy transfer of seismic waves separation of scattering effect from intrinsic attenuation II.Application of the theory to Hindu Kush region:Pure and Applied Geophysics.128.49 8O. Wu,R.S.,1 989,The perturbation method in elastic wave scattering:Pure and Applied Geophysics,131,605 637. Wu,R.S. 1 994、Wide—angle elastic wave one—way propagation in heterogene0us media and an elastic wave complex—screen method:Journal of Geophysics Research.4 75 1—766. Xi,X.,and Yao,Y.,2002,Simulations of random medium model and intermixed random medium:Earth Science— Journal of China University of Geosciences(in Chinese), 27.67—71. Liu Jiong See biography and photo in the Applied Geophysics September 2009 issue,P 270. 摘要:单程波算子的可分近似形式是关于空间变量 和波数变量的函数,这使得快速傅里叶变换的利用 成为可能,从而极大提高了计算效率。从函数近似 的角度来看,最优可分近似方法与其他几种可分近 似方法一样都具有单程波算子可分表达的形式,但 最优可分近似方法是其中唯一的函数整体近似的方 法,这就造成了该方法在相位误差曲线、脉冲响应、 模型偏移结果上表现出与其他可分近似方法不一样 的特征:该方法精度较高,且随着阶数的增大,对 速度变化的敏感度降低。 关键词:单程波算子,偏移,可分近似,最优可分 近似 高精度频率衰减分析技术及其应用//High—precision frequency attenuation analysis and its application, 熊 晓军・,贺锡雷t,蒲勇2贺振华 ,林凯 ,Applied Geophysics,8(4),P.337-343 (1.成都理工大学地球探测与信息技术教育部重点实 验室,成都,610059;2.中石化勘探南方分公司,成都, 610041) 摘要:本文从含流体孔隙介质中的地震波场的衰减 理论出发,对常规的频率衰减分析技术中的“低频 阴影”和“频率衰减梯度”分析方法进行了改进, 提出了一种高精度的频率衰减分析技术。首先,通 过引入三参数小波变换和时频聚焦准则,发展了一 种基于自适应三参数小波变换的高精度时频分析方 法,其不仅具有很高的时一频分辨率(有利于“低 频阴影”分析),而且其频谱只有一个峰,旁瓣比较 小(有利于“频率衰减梯度”分析)。其次,采用基 于最小二乘法的Nelder—Mead非线性算法对频谱的 衰减部分进行拟合计算,可以准确地计算衰减系数, 提高了“频率衰减梯度”的计算精度。实际资料的 计算结果表明,本文提出的综合“低频阴影”和“频 率衰减梯度”方法的频率衰减分析技术能够有效地 圈定碳酸盐岩鲕滩储层的发育区域,且两种方法具 有很好的一致性,有效地提高了储层预测的可靠性, 从而降低了勘探风险。 关键词:衰减分析,低频阴影,频率衰减梯度,时 频分析 随机弹性介质中地震波散射衰减分析//An analysis of seismic scattering attenuation in a random elastic medium,刘炯1,2,3魏修成 ,季玉新 -一,陈天胜1,2刘 春园 ,张春涛 ,戴明刚2 Applied Geophysics,8(4), P.344.354 (1.中国石化多波地震技术重点实验室,北京10083; 2.中国石化石油勘探开发研究院,北京100083;3.中 372 国石油大学(北京)地球物理与信息工程学院,北 京102249) 摘要:地震波衰减一直是许多学科研究的热点,因为 可以反映介质的特性。导致地震波衰减的因素很多, 如:传播过程中由于能量扩散导致的几何衰减,固 体岩石内部晶粒问相对滑移导致的摩擦衰减,岩石 结构不均匀引起的地震波散射衰减。本文主要从统 计的观点出发,通过多次数值模拟的方法研究纵波 散射在随机弹性介质中所引发的衰减。首先用随机 理论建立了二维空间随机弹性介质模型,然后用错 格伪谱法的数值方法模拟了波在随机介质中的传播, 再通过波场中虚拟检波器的记录,用谱比法估计了 弹性波在随机介质中的散射衰减。不同非均匀程度 随机弹性介质中的数值结果表明:介质不均匀程度 越高,散射衰减越大;在散射体尺寸小于波长的前 提下,不同散射体尺寸的计算结果说明:散射体尺 寸越大,弹性波衰减越明显。最后提出了一种不均 匀孔隙介质中流体流动衰减的方法。通过对随机孔 隙介质中地震波的总衰减和散射衰减分别进行了计 算,并定量得出了随机孔隙介质中流体流动衰减, 结果表明:在实际地震频段下,当介质不均匀尺度 101米量级时,散射衰减比流体流动衰减要大,散射 衰减是地震波在实际不均匀岩石孔隙介质中衰减的 主要原因。 关键词:散射衰减,随机弹性介质,谱比法 基于方向性滤波的地质体突出显示//Highlighting display of geologic body based on directivity filtering, 文晓涛・u一,贺振华2 7黄德济,,陈学华・r,Applied Geophysics,8(4),P.355-362 (1.成都理工大学油气藏地质及开发工程国家重点实 验室,四川成都,610059;2.成都理工大学地球物 理学院,四川成都6100059) 摘要:为了突出地质体的轮廓并保证横向分辨率, 在进行滤波的同时必须保证地质体的边缘。为此, 引入了各向异性扩散滤波技术。该技术的核心思想 是沿着边缘的方向扩散滤波,而垂直边缘的方向则 不扩散,这样就保证了滤波具有方向性。为了增强 该技术的实用性,对各向异性扩散模型的参数进行 了讨论,在充分考虑地震分辨率的情况下,给出扩 散系数公式,推导分析了扩散门限参数的选取原则。 依据本文提出的公式及原则,对XX区的潮道、YY 区的砂体进行了突出显示,良好的应用效果证实了 本文提出的参数选取原则是合理的。