GEOPHYSICS, VOL. 64, NO.4 (JULY-AUGUST 1999); P.1172-1180, 15 FIGS., I TABLE. Analysis of multiazimuthal VSP data for anisotropy and AVO W. Scott Leaney *, Colin M. Sayers, and Douglas E. Miller ** ABSTRACT Multioffset vertical seismic profile (VSP) experiments, commonly referred to as walkaways, enable anisotropy to be measured reliably in the field. The results can be fed into modeling programs to study the impact of anisotropy on velocity analysis, migration, and ampli- tude versus offset (AVO). Properly designed multioffset VSPs can also provide the target AVO response mea- sured under optimum conditions, since the wavelet is recorded just above the reflectors of interest with mini- mal reflection point dispersal. In this paper, the multioffset VSP technique is ex- tended to include multioffset azimuths, and a multiaz- imuthal multiple VSP data set acquired over a carbonate reservoir is analyzed for P-wave anisotropy and AVO. Direct arrival times down to the overlying shale and re- flection times and amplitudes from the carbonate are analyzed. Data analysis involves a three-term fit to ac- count for nonhyperbolic moveout, dip, and azimuthal INTRODUCTION The geometry of marine multioffset vertical seismic profile (VSP) experiments, commonly referred to as walkaways, ac- quired with an array of three-component receivers downhole, is ideally suited to the task of measuring angle-dependent phe- nomena associated with elastic wave propagation. For example, a method of estimating anisotropy parameters for transversely isotropic (TI) media from phase slowness data was introduced by Gaiser (1990). In that method, vertical and horizontal phase slownesses are inverted for TI moduli under the assumption of weak anisotropy. Miller and Spencer (1994) showed that P-wave phase-slowness data could be inverted exactly for TI moduli without any assumption of weak anisotropy, provided an estimate of the vertical shear velocity is available. This method was first applied to a real multioffset VSP data set by Miller et al. (1994). anisotropy. Results indicate that the overlying shale is transversely isotropic with a vertical axis of symme- try (VTI), while the carbonate shows 4-5% azimuthal anisotropy in traveltimes. The fast direction is consistent with the maximum horizontal stress orientation deter- mined from break-out logs and is also consistent with the strike of major faults. AVO analysis of the reflection from the top of the carbonate layer shows a critical angle re- duction in the fast direction and maximum gradient in the slow direction. This agrees with modeling and indicates a greater amplitude sensitivity in the slow direction— the direction perpendicular to fracture strike. In princi- ple, 3-D surveys should have wide azimuthal coverage to characterize fractured reservoirs. If this is not possible, it is important to have azimuthal line coverage in the min- imum horizontal stress direction to optimize the use of AVO for fractured reservoir characterization. This direc- tion can be obtained from multiazimuthal walkaways us- ing the azimuthal P-wave analysis techniques presented. Miller and Spencer (1994) also studied the problem of esti- mating the moduli of an orthorhombic medium consisting of a set of vertical fractures in a TI medium with a vertical axis of symmetry; they found that if the symmetry planes are known in advance, a minimum data set would involve walkaway ex- periments in the two symmetry planes plus a third at an inter- mediate azimuth. The approach of Miller and Spencer (1994) to orthorhombic moduli inversion from multiazimuthal walk- away data was reviewed in Leaney et al. (1995). Sayers (1997) showed how the TI inversion method of Miller and Spencer (1994) could be used in the presence of moderate regional dip. Multioffset VSP data also have been used to measure the AVO response of a reservoir. This was first done with land data by Coulombe et al. (1991). Henderson et al. (1994) and Smith et al. (1995) used marine multioffset VSP data to cal- ibrate surface gathers. Breton and Cadoret (1997) discussed Manuscript received by the Editor March 4,1998; revised manuscript received December 28, 1998. *Schlumberger House, Gatwick Airport, West Sussex RH6 ONZ, U.K.; E-mail: leaney@gatwick.geco-prakla.slb.com. Formerly Schlumberger Cambridge Research, Madingley Road, Cambridge CB3 OEL, U.K.; presently Schlumberger Geco-Prakla, 1325 S. Dairy Ashford, Houston, TX 77077; E-mail: sayers@houston.geco-prakla.slb.com. **Schlumberger-Doll Research, Old Quarry Rd., Ridgefield, CT 06877-4108; E-mail: miller@ridgefield.sdr.slb.com. © 1999 Society of Exploration Geophysicists. All rights reserved. 1172 Downloaded 25 Dec 2010 to 199.6.131.16. Redistribution subject to SEG license or copyright; see Terms of Use at http://segdl.org/