Preprint

UCRL-JC-137646

3-D elastic numericalmodeling of a complex saltstructure

L. House, S. Larsen and J.B. Bednar

This article was submitted toSociety of Exploration Geophysicists 70’h Annual Meeting,Calgary, Canada, August 6-11,2000

February 17,2000U.S. Depmtment of Energy

&

LawrenceLivennoreNationalLaboratory

Approved for public release; further dissemination unlimited

DISCLAIMER

This document was prepared as an account of work sponsored by an agency of the United StatesGovernment. Neither the United States Government nor the University of California nor any of theiremployees, makes any warranty, express or implied, or assumes any legal liability or responsibility forthe accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, orrepresents that its use would not infringe privately owned rights. Reference herein to any specificcommercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does notnecessarily constitute or imply its endorsement, recommendation, or favoring by the United StatesGovernment or the University of California. The views and opinions of authors expressed herein do notnecessarily state or reflect those of the United States Government or the University of California, andshall not be used for advertising or product endorsement purposes.

This is a preprint of a paper intended for publication in a journal or proceedings. Since changes maybemade before publication, this preprint is made available with the understanding that it will not be citedor reproduced without the permission of the author.

This report has been reproduceddirectly from the best available copy.

Available to DOE and 00E contractors from theOffice of Scientific and Technical Information

P.O. Box 62, Oak Ridge, TN 37831priCes available from (423) 576.8401

ht@/ /apollo.osti.gov/bridge/

Available to the public from theNational Tectilcal Information Service

U.S. Department of Commerce5285 Port Royal Rd.,

Springfield, VA 22161http: //www.ntis.gov/

OR

Lawrence Livermore National LaboratoqTechnical Information Department’s Digital Libraxy

http: //www.llnl.gov/tid /Library.html

3-D elastic numerical modeling of a complex salt structure12igh House*, L.QsAlantos Seismic Research CenterLos Akunos National Laboratory, Shawn Larsen, LawrenceLiverntore National Laboratory, and J. Bee Bednar, Advanced Data Solutions

s~

Reliably prwessing, imaging, and interpreting wismic datafrom areas with complicated structures, such as sub-salt,requires a thorough understanding of elastic as well as acotts-tic wave propagation. Elastic numerical modeling is anessential mol to develop that understanding. While 2-D elas-tic modeling is in common use, 3-D elastic mwleling hasbeen too computationally intensive to be used rmtinely,Recent advances in computing hardware, includingcommodity-based hardware, have substantially reduced com-puting costs. These advances are making 3-D elasdcnumerical mtiling more feasible. A series of example 3-Delastic calculations were performed using a implicatedsb’uctute, the SEG/EAGE salt structure. The synthetic @acesshow that the effects of shear wave propagation can lx im-portant for imaging and intqretadon of images, and also forAVO and other applications that rely on trace amplitudes.Additional calculations are needed to L?etter identify andunderstand the complex wave propagation effects prcducedin complicated structures, such as the SEGiEAGE salt struc-ture.

Inttvtfuclion

There has b%n considerable rmmt intexest in undcmtandingelastic wave propagation and in best exploiting the infmma-tion that shear waves provide because of two recent devel-opments in exploration. First is increasing exploration ofareas with complicated structures and large velwitycontrasts, such as sub-salt. Second is the increasingavailability of multi-component seismic data

Qin, et al (1977) showed that even 2-D elastic simulationscould be a powerful tcml to correctly i&ntify shear (con-verted) waves. They also showed that shear waves provideinformation to mom reliably interpret seismic data frombelow salt. The greater reliability of their interpretationhelped reduce the risks of explorimg prospects below salt. Ifshear wave events ate not propaly identified, they can hemisleading artifacts and lead to erroneous intqnetations. Inaddition, since shear waves are more sensitive to the presenceof fluids than compressiotml waves are, shear wave data mayprovide more reliable estimates of reservoir properties. Thus,reservoirs can k better understood and more reliablymodeled and exploited.

Elastic wave modeling demands much more computerresources than are needed for acoustic wave modeling. Morememory is needed because of tfte greater number of model

parameters, and more compudng time is requkd to carry outthe computation. The additional variables inmduced bygoing to elastic computations add a factor of 3-1 to both thememory and computing time needed. But, tiis factor alone isnot the most important impact that mstdts from going fromacoustic to elasdc simtdations. ?lte fact that shear wavevelocities are lower than compressional wave velocities andtheir wavelengths are proportionally shorter means that elas-tic wave simulations generally require that the velccity modelhe spxifiexf on a freer spatial grid interval. That furtAerincreases the memory and computing dtne needed. The com-bined result of the additiond elastic variables and the finergrid spacing required by elastic modeling is an overallin- in computing time by 25-lfN3 times compamd toacoustic mcdel calculations.

The SEG/EAGE 3-D numerical modeling project (see Amin-mdeh, et al, 1997) demonstrated that numerical modelingcould be an important tool for tssting and validating imagingmethcds. That pmjsct &signed model suttctttres to mimiccomplex, yet plausible, salt and overthrust geologic struc-tures. Millions of simulated seismic traces were computed inthese structures, in a tour de force of computing that har-nessed the combined power of supmmmputem in the US.and Europe. The result was a muld-terabyte data set that isavailable to researchers world-wide.

As powerful as the computing resources used by theSEGIEAGE numerical mcdeling project were, however, theproject was only able to carry out acoustic wave calculations.The project was obliged to choose to carry out acousdcsimulations because of the relatively high cost of the com-puting resources that were available at the time (about 5 yearsago). Dramatic recent incn%mes in computing resoumes,notably memory, computing speed, and storage space, makeelastic wave computations, even in 3-D, more feasible.Development of computing clusters based on commcdity-PC’s has dramatically accelerated the rates of increase incomputing capability and of decre= in cost. These trendsate likely m continue, and will allow 3-D elasdc wave calcu-lations to ix used more mudnely, even by relatively smallorganizations.

To help emphasize the benefits of 3-D elastic wave simtda-tions, we show the results of example calctdadons done in theSEG/EAGE salt mcdel.

Method

3-D elasdc computations were carried out using the modeling

3-D dastic numerical modeling

Ua mm ,CM

7 1

Acoustic, P Comp E188UC,P Comp Elastic,Y tip

IFigure 3. Shot gathers from shot 145, acoustic and elastic calculations. Plots show near 60 traces from a portion of one cableof a K ocean bottom wdde survev. Addidonal details are in the text H@oDhone fP wave) maces from acoustic and elastic

I calculationsare at the left and ce&r, respectively,and transversegeophone traces from elmtic calculations am at tbe right.Labels for reflection events are described in the text.

1ix a PSSSSP event in the nomenclature used earlier. Neither and clusters, will help to show the relative costs and benefitsof these converted wave events (PSSP or PSSSSP) can beseen on the hydmpbone gather. There m many smaller, lesscoherent events in the lower 2 s portion of the mmsversegeophone traces that may result from multi-pathing withinthe salt WY. Both the top and bottom surfaces of tie salt intie mtiel are it’regulm, and could prduce small amplitudeevents that are difficult to interpret. There is a smallamplitude event neat the bottom right of the piot that ha?,inverse moveout. This may be red, although there is noobvious feature in the velocity smucture that could prcdtcesuch an event at that time, so it could be a computationalartifact.

conclusions

3-D elastic numerical calculations fmm the SEGIEAGE saltstructureshow the impommce of mcdding to understandingelastic wave propagation effects in a complicated structure.They also illustrate the greater complexity of seismic eventsthat aIe likely to he in real seismic data fmm complexstructures with large velocity contrasts. These complexitiescreate difficulties for botb interpretation and routineprocessing and imaging, as they result from combined Pwave and S wave propagation and interactions with the 3-Dgmmeuy of the salt bcdy in this model. The elastic wavecalculations also show different amplitudes as a function ofoffset.

More computationsare neededto fully understand tie effectof elastic wave interactions with complicated structures. Theexamples shown here were computed on one system;additional computations 0. other types of systems, Mb MPP

of different solutions to the computing problem.

Acknowledgments

Leon Thomsen, and Walter Kessinger provided helpfuladvice and discussions. This work was performed under theauspices of tbe U.S. Department of Energy by the Universityof California, Los Akitnos National Laboratory, undercontract No. W-7045 -ENG-36, and Lmwence LivemmreNadonal Laixmtoty, under contract w-7045-ENG 4S.

References

Aminzadeh, F., J. Brat, and T. Kunz, 1997, 3-D satt andoverthntst mcdels, SEG03AGE 3-D Mc4eling Series, No. 1,SIX. Expl. Geophys. Ttds% OK.

Gardner, G.H.F., L.W. Gardner, and AR. Gregory, 1974,Formation velccity and density - the diagnostic basis forstratigraphlc traps, Geophysics, 39,770-780.

Larsen, S., and J. Grieger, 1998, Elastic modeling initiative,part 111 3-D computational mcdeling, Expanded Abstracts,68th Amwd SEG meeting, p 1803-1806.

Qin, F., 1.P. Leveille, 1.A. Weigant, S.N. Ciwkles, B.K.Boslaugh, J, McGinnis, H.S. Sanchez, and J.B. Bednar, 1997,Application of elastic modeling to subsalt interpretaticm,Expanded Abstracts, 67th AMual SEG meeting, p 1092-1095,

.— ——