Acquisition footprint is a pattern on seismic data which is mainly caused by the sparseness of the acquisition geometry. It hinders seismic interpreters from accurate structure analysis by masking geological significances such as fault lineaments, channels and karsts. It is triggered by coherent changes in case of variable acquisition parameters and/or methods in the field, or alignments caused by the direction of data acquisition, and it is predominant on sparse acquisition geometries. Furthermore, it might be accentuated by multi-channels seismic processing such as stacking, DMO and pre-stack time migration if the aliased noise produced by the sparse geometry does creep into the seismic data.
In general, the acquisition footprint signature is rather strong in the shallow events due to relatively lower fold. Hence, it is strongly dependent on trace mute designs and sensitive to the amount of moveout and NMO stretch, where the influence is progressively healed with depth due to higher fold, wider mute parameters and smaller NMO shift.
It is known that remnant acquisition footprints often exist on the data after the application of the 3D F-Kx-Ky wave number notch filter. Although it may not always appear on the vertical cross-sections and time slices prominently after the attenuation, it might be pronounced on edge detection attributes, such as coherency cube, curvature and/or seismic impedance. There is some ongoing debate in Abu Dhabi between processing geophysicists and interpretation geophysicists as to whether apply harsh filter to clean up the residual footprint or to be more conservative while preserving some desirable subtle geological features as possible.
This paper advocates the necessity of time-variant 3D notch filtering approach regardless of the acquisition geometry and/or the area by showing some case histories from 3D seismic surveys offshore and onshore Abu Dhabi with three different acquisition geometries; orthogonal OBC patch layout, orthogonal land layout, and parallel OBC patch layout. Change of footprint characteristics in time, in association with adaptiveness of various notch filters in temporal/spatial domain, will b e discussed and demonstrated for each geometry type.
