الفهرس 
Pre-Processing StageOnly 14 pages are availabe for public view
 |  | from | 298 |  |  |
| | | from | 298 | | |
Abstract
The study area is located in the western offshore Egyptian province which exists west of the prolific Nile Delta Cone. It covers an area of approximately 82,450 Km2 that is equivalent to 46% of the whole offshore Egyptian Mediterranean. The area of study covers from very shallow to ultra-deep water depth, with a majority of very deep water province.
To date, exploration in this large area has been very limited, and the quality of the available seismic data is poor, consequently the petroleum systems are currently not well understood. Only two dry exploration wells have been drilled.
The main purpose of the study is to reprocess the available 2D seismic data in the area using broadband processing sequence to generate high quality 2D seismic data. The best practice noise attenuation and de-multiple techniques, together with high precision velocity picking and imaging are used to achieve a clear and high resolution image of the subsurface geology. In order to achieve this target, the current study started with the description of the geology of the area, including a detailed discussion of the stratigraphic rock units, structures and tectonic evaluation of the offshore Egyptian Mediterranean, where the study area is located.
The available Magnetic data has been used to delineate the depth of the Basement to be used later in the migration velocity model building.
The processing flow used to achieve the objective of this study has been divided into three main stages “Pre-processing”, “De-multiple and Offset Processing” and “Migration and Post-processing”.
The Pre-processing stage was concerned mainly with removing the existing noise, removing all types of ghosts, zero phasing the data and precondition the data to the de-multiple stage.
Data was contaminated with various coherent and non-coherent types of noise. A dedicated de-noise approach was employed in various domains to deal with the different types of noise, while maximizing the preservation of the primary signal. Most of the swell noise, direct arrival, seismic interference and linear noise have been attenuated properly without any noticeable primary leakage.
De-ghosting was very challenging in this study, and it was one of the key reasons behind the tremendous uplift obtained in the final data, it has been applied in a 2-D sense by taking account of the emergence angle. The solution is deterministic using the measured receiver/source depths. Zero phasing the data has been done using the best available wavelet representing each survey. The derivation of the filter has been done using “BOSS+” software.
The errors introduced by the forward motion of the receivers and the source due to the transit of the vessel during acquisition has been removed by the application of the Receiver Motion Correction. In addition to the Tidal statics which helps to refer all the heights to the mean sea level as a reference datum.
Surface multiples together with internal multiples offered a big challenge in this data. Multiple contamination was dealt with using a cascaded de-multiple approach, starting with 2D SRME and 2D Wave Extrapolation, followed by a mild Hi-resolution Radon de-multiple application pre- migration, and ending with a more tuned Hi-resolution Radon de-multiple after migration. Considerable time and effort was invested in the proper testing of 2D SRME & 2D Wave Extrapolation methods, as well as tuning the multi-model adaptive subtraction of the modelled multiples from the data, to ensure optimum subtraction of multiples, while maximizing the preservation of primary data.
Data was still contaminated with residual coherent and non-coherent types of noise. A rigorous de-noise approach was employed in various domains to deal with the different types of residual noise. A lot of effort was invested in the proper application of multiple diffraction removal (MDR) technique in order to attenuate the strong residual diffracted multiples that offered a very big challenge for all 2D lines.
Velocity picking offered another big challenge for this data set due to the sensitivity of the data to small changes in velocity. This was handled through various runs of manual velocity picking together with the use of high density automated velocity picking using PGS’s proprietary software “optivel”. The big effort made in producing the various manual velocity fields, offered a very good reference field for the Auto-picking, besides the approximate basement depth calculated from the magnetic interpretation which also contributed to get a good quality of the manual velocity model. Then an intensive workflow has been followed to get the final smoothed migration velocity model.
Anisotropic Kirchhoff Pre-Stack Time Migration was used to get the final migrated results, which were way better than the quality of legacy data. The main uplift observed in the raw migration results was due to the cleaner input after the application of all the previously mentioned de-noise and de-multiple flows and also the very accurate velocity model.
Post Processing workflow has been applied to the migrated data with the hope to enhance the final imaging results even further. The auto-picking has been used to pick the final stacking velocity and final tune the existing migration velocity fields, both RMS velocity and ETA fields. Also Radon de-multiple and Curvelet transformation was used to remove any residual existing noise and multiple before stacking the data. Trim statics application was needed in order to correct for any small-scale non flat horizon before stacking. The final full stack angle range were chosen to be 7-30 degree which prove to be the best angle range to stack the data and getting rid of the near offset multiple and far offset stretching. Following to the stacking, the Q compensation has been applied “Amplitude factor only” to enhance temporal and spatial resolution in addition to the Time Variant Filter “TVF “and Time Variant Gain “TVG” to get the final stack image of the data.
For the final assessment of achieving the study objective, a comparison between the old legacy data ”old processing” which was available with EGAS for any future exploration activity and the new processing results has been presented in Chapter 6, section 6.3.
Reviewing the seven Figures in section 6.3, it can be observed that the final new PSTM re-processing results are superior to Legacy final PSTM results in following:
– Superior enhancement in spectral content leading to broadband data, without boosting noise.
– Superior improvement in de-multiple results, leading to a far better delineation of deep geology.
– Superior imaging of shallow faults and deep geology.
– Clear improvement in lateral and temporal resolution of the data.
The new results were a game changer in the Egyptian Mediterranean province, the tremendous uplift which has been achieved in this study revealed a lot of the hidden details of this province, which was kept unexplored for long time. Now, there is a huge interest from the International exploration companies about this study area.