Angle Domain Common Image Gathers And Migration Velocity Analysis

The prestack depth migration develops fast as a result of the promotion of oiland gas exploration and complicated exploration aims. Wave equation prestack depthmigration is widely applied to the industrial production step by step. The prestackdepth migration requires relatively accurate input velocity model, so we starve for newmigration velocity analysis(MVA) methods which suit wave equation prestack depthmigration. Recently, angle domain common image gathers (ADCIG) are introducedinto geophysics, they are considered to have great potential to both MVA and imaging.This paper intends to study the MVA methods using ADCIG.ADCIG reflect the seismic reflection amplitude versus incident angles. The pro-cess of extracting ADCIG is essentially to transform the multi-fold seismic data fromoffset domain to incident angle domain. Among the existing main methods of obtain-ing ADCIG, transforming offset domain common image gahers (ODCIG) into ADCIGby slant stack is more efficient, so this method is widely studied and applied. In in-cident angle domain, seismic data have outstanding properties. ADCIG gained fromwave equation migration are proved to be free of kinematic artifacts. This is the max-imal merit apart from other kinds of CIG. It is also the main reason why ADCIG arecurrently considered as the most reasonable CIG.When the migration velocity equals to medium velocity, the event of ADCIGis fiat, and angles in ADCIG are true seismic incident angles. Otherwise, the eventcouldn't be fiat. It will be curved upward or downward. The angles won't be the trueincident angles any more, they are apparent incident angles, If the migration velocity ishigher than actual velocity, the apparent angles are greater than those of the trues, andvice versa. In ADCIG, the residual moveout (RMO) with a positive migration velocityerror is larger than that with a negative one of the same factor, that is to say, the RMOof ADCIG is more sensitive to the positive migration velocity error. This property sug-gests that taking large migration velocity as the initial one for velocity analysis availsto judge the velocity error.In this dissertation, we obtain time-shift offset domain common image gathers (TSODCIG) and time-shift angle domain common image gathers (TSADCIG) by in-troducing the space-time-shift imaging condition. Both ADCIG imaging condition andtime-shift imaging condition could be considered as special cases of space-time-shiftimaging condition. At the focal depth, the wavefield energy of time-shift commonimage gathers (TSCIG) is the largest, and the event of ADCIG of TSADCIG is flat.On TSADCIG, we can accurately locate the focal depth by locating the flat gathers.Whether the migration velocity is correct, we can always find a certain time-shift, theevent of its corresponding ADCIG is flat. Based on this idea, this paper unifies thevelocity criterions for the two main MVA methods, namely, the flat-event criterion ofADCIG used in the residual curvature analysis (RCA) are equivalent to the equalityof imaging depth and focal depth using in the depth focusing analysis (DFA). The flatevent of ADCIG is the inevitable result of focused wavefield. The RCA and the DFA,essentially, both pursue the coherence of the imaging depth and focal depth.The ultimate goal of this research is to get the velocity structure of subsurfacemedium. The in-depth analysis about the RMO of TSADCIG results in the, updatingformulae for MVA. Combining the advantage of DFA and RCA, the author deviseda velocity analysis method which is based on TSADCIG/TSCIG. On TSADCIG, thefocal depth corresponds to the flat event, and on TSCIG, it can be found by lookingfor the maximum focusing energy. These two methods supplement each other and bothparticipate in estimating migration velocity errors. Tests on synthetic seismic data andreal field data processing show us the validity of this analysis method.