Data Dependent Adaptive Deghosting - Application to Vintage Data

While broadband acquisition techniques such as multi measurement streamers provide superior solutions for broadband data, there has been a growing interest in recovering those lost frequencies from legacy seismic data that were not acquired with a broadband acquisition technique. For example, legacy data were often acquired using a shallow receiver and source depth to deliberately place the first non-zero destructive interference pressure notch outside the range of signal frequencies. Although in this situation the source and receiver ghost notches are at the ends of the seismic spectrum, reprocessing the data with sophisticated deghosting methods can still help to extend the bandwidth in this case. We applied an adaptive source and receiver deghosting to a vintage Southern North Sea dataset to increase the temporal bandwidth. The deghosting method attempts to take into account crossline wave propagation subject to the inherent limitations of the single-streamer single-component geometry. It is data-dependent, and tolerates larger errors in water velocity or depth of the source and receivers than deterministic methods. A sparsity constraint in the algorithm enhances the low frequencies and hence the seismic inversion results. Low-frequency interpretation and seismic imaging of steep dips are seen to be improved using current methodology.