Overcoming the challenges of a shallow-water sparse wide-azimuth survey to improve deep reservoir imaging in the East China Sea

Abstract

A new broadband wide-azimuth towed-streamer (WATS) survey was acquired to better resolve reservoir compartments in a shallow-water region of the East China Sea. To o\x8fset the shortcomings of narrow-azimuth acquisition along the strike direction, two vessels were added side-by-side as additional source vessels to form the WATS acquisition geometry for this survey. is WATS acquisition was much sparser than typical WATS surveys used in deepwater environments due to its onesided con\xadguration. e combination of sparse acquisition, shallow water, and deep targets set the challenge of how to optimally reveal the potential of side-gun data to improve the \xadnal image. ree-dimensional e\x8fects and severe aliasing in the crossline direction pose signi\xadcant challenges for side-gun data processing. We present a comprehensive work\x80ow to resolve these challenges consisting of 3D deghosting, 3D model-based water-layer demultiple, 3D surface-related multiple elimination, and 4D regularization for sparse and shallow-water wide-azimuth data. A tilted orthorhombic velocity model is built with better constraints from the wide-azimuth data, leading to improved fault positioning and imaging. Side-gun data clearly enhance the \xadnal target reservoir image and tie better with well data due to improved illumination. A new channel is discovered based on interpretation from the inverted VP /VS, explaining the previous incorrect prediction for one failed well that was drilled into a thinner and shallower channel unconnected to the main reservoir. An analysis of the impact of side-gun data from di\x8ferent o\x8fsets and azimuths shows that better azimuthal distribution within middle o\x8fset ranges had a more signi\xadcant impact than far o\x8fsets in the \xadnal image of this survey. is information provides valuable reference in similar geologic conditions for future acquisition designs. Introduction e study area is located in a shallow-water region of the East China Sea with a water bottom around 70–100 m in depth. e production \xadeld has several wells drilled into the target reservoir. One was a failed well, while the rest were all successful and drilled into the production \xadeld. e legacy seismic image indicates that all of the wells, including the failed one, should have been drilled into the same structure between 3.1 and 3.6 km in depth. erefore, it was concluded that the legacy data were inadequate for resolving reservoir compartments. New broadband Yun Wei1, Hua Chen1, Senqing Hu1, Peipei Deng2, Yongdeng Xiao2, Srujan Poonamalli2, Robert To2, Joe Zhou2, Jason Sun2, Gang Yao1, and Yu Jiang1 data with good low-frequency signals were preferred (Chen et al., 2017). Legacy acquisition followed the dip direction, but the new acquisition had to follow the strike direction due to operational limitations as shown in Figure 1a. Meanwhile, 3D e\x8fects of the structure favor more azimuthal information. To overcome this limitation, two vessels were added side-by-side as additional source vessels on one side of the streamer vessel to form the wide-azimuth towed-streamer (WATS) acquisition geometry. e side guns provided the required re\x80ection information in the dip direction. e WATS acquisition setup consisted of three vessels and four guns shooting in sequence, as shown in Figure 1b, with the acquisition carried out in two passes. e streamer boat had two guns and 10 cables with a 100 m cable separation and 6 km cable length. e shot geometry had a 50 m interval between the two guns and a 50 m interval between shot points from the same gun. e cable pro\xadle was slanted for ghost notch diversity and broader bandwidth with depth varying from 7 m at near o\x8fsets to 40 m at far o\x8fsets. e side vessels each had one gun and were positioned laterally at a distance of 1 and 2 km from the streamer vessel centerline for the \xadrst pass and at a distance of 3 and 4 km for the second pass. e streamer vessel centerline shifted 12.5 m along the crossline direction during the second pass to improve subsurface coverage. Rather than at the center of the cable where the vessel would be within 10° of the nearest cable feathering range, as shown in Figure 1b, the inner side vessel was located 2 km behind the \xadrst channel in the inline direction in order to avoid potential entanglement from cable feathering. e outer side vessel was located 3 km behind the \xadrst channel. Figure 1c shows the azimuth and o\x8fset distribution of this acquisition geometry. Relative to typical WATS surveys in deepwater environments such as the Gulf of Mexico (Michell et al., 2006), the new acquisition was sparser for the following two reasons. Only one streamer vessel was deployed in the acquisition, and the side vessels were positioned on only one side of the streamer vessel. With such sparsity of the wide-azimuth acquisition in a shallow-water environment, the key challenge is how to leverage the full potential of the additional source vessels by data processing. Among the many challenges that must be overcome are deghosting, demultiple, regularization, and anisotropic velocity model building. In the following sections, we discuss these challenges and demonstrate the full advantages of this new WATS acquisition. 1CNOOC Shanghai, Shanghai, China. E-mail: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]. 2CGG, Singapore. E-mail: [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]. https://doi.org/10.1190/tle38080610.1.