Comparing Seismic Resolution and Signal and Noise Quality between Dense Point-receiver and Conventional Data over the Bakken Formation in North Dakota
Banik, Niranjan C.; Salama, Antoun; Egan, Mark; Koesoemadinata, Adam; El-Kaseeh, Khadir G.
Substantial vertical and spatial heterogeneities in reservoir and completion qualities are now recognized characteristics of shale reservoirs. This is especially true in the Bakken resource of the Williston Basin, North Dakota, in which the production comes mainly from the thin Middle Bakken formation. The lithology of the Bakken changes rapidly both vertically and spatially. Tracking the Middle Bakken is a major challenge for drillers. Improving seismic resolution to better image this complex reservoir is a key objective of Bakken operators. Because resolution is interplay of both vertical and spatial samplings, and intra-array statics plays an important role in resolution, both must be considered in improving the resolution. Conventional seismic acquisition and processing focuses on vertical sampling and ignores the spatial sampling. This puts the conventional system at an inherent disadvantage with respect to the point-receiver acquisition and processing system in resolving thin beds. This is exemplified in the present paper through processing a dense point-receiver data set acquired over the Williston Basin in North Dakota, where the exploration and production target is the Mid Bakken.
Two specific processing methods were applied to the data set. In one method we addressed the noise reduction and statics correction in the point-receiver domain and in the other method, we performed digital array-forming and simulation of conventional acquisition and then applied noise reduction and statics application. This enabled us to compare the signal and noise quality of dense point-receiver data with that of the conventional data over the same area and using the same acquisition and processing methods.
We find significant improvement in signal-to-noise ratio and usable bandwidth in the point-receiver data compared to that in the conventional data. The ability of statics correction before array-forming helps significantly in this regard. The large channel count in point-receiver data also helps in reducing certain types of noise in the data. The improvement in signal-to-noise ratio and usable band width are reflected in the final acoustic impedance inversion results.
The high signal and noise ratio in dense point-receiver data helps whiten the spectral bandwidth further to resolve the Middle Bakken in the area, maintaining the spatial fidelity of data as well. These are evident in well-ties and seismic inversion results of the two data sets.
AAPG Search and Discovery Article #90163©2013AAPG 2013 Annual Convention and Exhibition, Pittsburgh, Pennsylvania, May 19-22, 2013