Southern North Sea preSDM imaging usinggridded tomography

摘要

Exploration in the Southern North Sea using conventional imaging techniques is hampered byrncomplexities in the Mesozoic overburden and the Zechstein evaporites with dolomitic raftingrnoverlying potential targets. 3D preSDM imaging has come into widespread use in recent years, in anrnattempt to resolve such problems.rnHere we present a case history from the GdF acreage over Quad 43 and 44, where an eight surveyrnmerge covering some 1500 sq.km was reprocessed to yield a coherent single input volume for bothrnpreSTM and preSDM imaging. One of the surveys was newly acquired using longer offsets, whichrnhad a noticeable impact on velocity estimation and signal quality.rnCurrent practice in velocity model building usually resorts to one of two approaches: the layer-basedrnand the gridded.rnThe layer-based approach has hitherto been commonly used for North Sea type environments, wherernsedimentary interfaces delimit changes in the velocity field and the geology 'lends itself' to a layerbasedrnmodel representation. In other words, we encourage preconceived bias, as we consider it to be arnmeaningful geological constraint on the solution.rnThe gridded approach is usually adopted in environments where the velocity regime is decoupled fromrnthe sedimentation, and is governed primarily by vertical compaction gradients (velocity increasingrnwith depth), controlled by de-watering, with iso-velocity contours sub-paralleling the sea bed.rnHowever, recent work conducted in complex Mesozoic Chalk environments in the North Sea hasrndemonstrated the advantages of removing the constraints of a layer-based solution, so as to permit arngridded tomographic approach to uncover the more subtle changes associated with variable chalkrncompaction regimes. In addition, the rapid velocity variation associated with Zechstein rafting andrncomplex salt topography, which does not lend itself to a layered description, can also profit from arngridded representation.rnWith this in mind, we have opted here to use a gridded tomographic approach for the model building.rnAs with any tomographic solution, the key to success lies in having very dense picking of reliablerninput data, with appropriate constraints. We employed dense continuous automatic picking of residualrnmove-out in CRP gathers at each iteration, based on plane-wave destructors, followed by griddedrntomography, resulting in a smoothly varying velocity field. This approach offers the possibility ofrnquicker model update, as we do not need one preSDM iteration for each 'layer' in the model.rnThe resulting images (Figures 1 & 2) show good resolution of the Zechstein and Carboniferousrnsection, much improved on vintage processing. Comparisons of the preSTM and preSDM resultsrn(figure 3) demonstrate the benefit of depth imaging for fault positioning in the deeper section.rnObservations were also made on the possible effects of azimuthal anisotropy resulting from therndisparate data acquisition azimuths.