PERMFILE50081 - Laserfiche WebLink

Home Browse Search

Fig. 1 includes hro PSDM seismic images of the [tfahogany salt sill from a line shot close to the discovery well on Gulf of Mexico Ship Shoal Seuth Addition 61ock 3-k9. The sections-mie an automatic gain control (.AGC) section and the other a rela lice amplitude section-- show the salt to be about 3,500 ft thick a4 in the approximate location of the discocen• gall and as much .ts 5,000 ft thick nearby. A significant feature of the relative amplitude section (Fig. 16) is the ap- pearance of three strong events belot+' the salt sill. This shows that relative amplitude data can be preserved be- lo+v salt. I tq p l y pl i I I I I'~ I'~n q ~l I vni Ip I lpl ' II ~ ~ ~~ 1~ 1 Fig. 2 is article twill describe some of the' ocedures that have proven to ~' helpful in the production of image; such as these, with special attention to`. the pivotal step: development of ari`'. accurate velocity model. _ . What PSDM does - Full-volume 3D PSDM is an iterative process involving huge amounts of data. The input for a recent Diamond Geophysical survey over the Mahoga- ny structure exceeded 23 million traces. Computer time required to process that much data is extensit'e and costly. And the iterative nature of the proce- dures means the costs must be in- curred several times. Depth migration differs from time migration by accounting for lateral changes in the velocities at ~+hich sound travels through subsurface stra- ta. This is important in salt imaging because sound behaves muc}t differ- ently in salt than it does in sedi- ments-mainly traveling faster-and because salt shapes and depths are very irregular. Until 3D PSDM became practical, seismic events below the tops of salt formations were very difficult to mi- grate accurately because standard time-migration techniques could nut properly position salt tops on seismic sections. lNith the top of salt mi>ivsi- tioned, accurate migration is im~+ossi- ble for the bottom of salt and sedi- ments below. The first several iterations in 3D PSDM migrate data down to the top of salt. They ail essentially trial and er- ror migrations of 3D data based on assumptions about t+•here the top of salt occurs in depth and t•elocities of the salt and sediments above. The next iterations migrate data down to the base of salt. Subsequent iterations migrate data down through subsalt sediments. The process involves much inter- pretation. It works best-meaning it produces must aauracv with the fen•- est iterations-when the first iteration starts with accurate assumptions about t•elocities and salt shapes. This is possible onl}' ~rhen pnxess- ing and interpretation steps prior to the PSDD4 iterations-+cith data sill in the time domain-are taken with ex- treme care. In some cases, that means performing standard procedures in nonstandard ways. The effort must remain focusai on development of an accurate celr:it} model. The model derived from th PSD\t that produced the ima•~es i Fig. 1 is shott•n in Fig. ?. ;,.tvhi~ i _-;: ity~t 1 .che~ ;• D • ma and i Vii. , mn lan -anz ~ • :'less t i - ~- 7 e "fiel n ~ 'the ling 1 - :-,, OilBGas JOUrnal•Oct U,19940GJ SPECIAL -, ~:~.0 i.-:.ie I II 'I II~h I~