Seismic data acquisition using Ocean Bottom Seismic (OBS) nodes was conducted over the Deimos field in the Gulf of Mexico, in about 1000m of water. The field is located under a salt overhang. Wave equation modeling was used to determine the optimal location of the nodes. The ability to acquire data at very long offsets was particularly useful. Seismic operations proved to be flexible in an area with considerable surface and subsea infrastructure, operational interference and strong currents. As the market for OBS node technology develops and more surveys get acquired in the future, it is expected that operational best practices will lead to increased efficiency in the field. Pre-processing of the data included node positioning, timebreak alignment, and optimized wavefield separation and noise attenuation on vertical geophone data. Migration of the downgoing wavefield data proved to be successful down to the target depth. Processed results achieved to date show significant improvements in sub-salt imaging compared to existing narrow azimuth data.
In this complex structure land imaging case history, we observed lateral positioning differences beneath the dipping anisotropic overburden of about 120 m between the isotropic prestack time migration (PSTM) imaging and the anisotropic prestack depth migration (PSDM) imaging. After migrating the anisotropic PSDM with different velocity and anisotropy models, we estimated the lateral positioning uncertainty in the anisotropic depth image. In the area where we observed the largest positioning differences with the PSTM, we estimated the uncertainty due to anisotropy to be of the order of 30 m, and the uncertainty due to velocity to be of the order of 15 m. As the dip of the overburden decreased, the lateral positioning differences between the isotropic PSTM and the anisotropic PSDM decreased, as did the lateral positioning uncertainty of the anisotropic PSDM imaging.
The marine CSEM method is widely used in exploration as a method of de-risking prospects prior to drilling. However the method may also be applied to assist in appraisal of known reservoirs, and for monitoring changes in the properties of a reservoir during production. For these applications accurate, repeatable measurements are required. The repeatability of a CSEM survey will depend on a number of factors which must be characterized as accurately as possible. These factors can be divided into two groups: factors dependent on acquisition (for example uncertainties in source position or orientation), and environmental factors (for example seawater conductivity or the presence of shallow heterogeneity in the earth); the contribution of each depends on survey conditions and acquisition parameters. Since the total number of the parameters that could potentially affect the measured response is large, an important task is to restrict this number by identifying those that contribute most to survey uncertainty and repeatability. The purpose of this study is to produce a systematic review of the contribution of each potentially relevant factor, find the conditions affecting their impact and provide practical recommendations for performing repeatable surveys in the field.
Seismic data acquisition using autonomous OBS nodes was conducted in about 1000 m of water in the Gulf of Mexico. After the initial acquisition, nodes were redeployed and three repeat acquisition phases were performed over a period of 60 days. In the first phase, some nodes were collocated to establish best possible repeatability with common shots. Final node positions were determined by first arrival time analysis.
The most significant factors affecting node data repeatability were found to be the scattered shear noise sensitivity to small changes in node position, the variation of the seismic wave velocity in the water column and the effect of strong loop currents on the operational ability to reproduce source positions. Data processing to mitigate the noise variations and comprehending the water column velocity variations in the depth migration step produced 2D image difference sections with an average NRMS ~ 10% when source positions were well matched, and NRMS ~ 20% when they were not.