Source De-ghosting and Directional Designature Using Near-field Derived Airgun Signatures

We have shown in previous work that directional airgun far-field signatures can be derived from measurements made near to the guns through an inversion process, as an alternative to the commonly used iterative method. This approach has the advantage of being more accurate at low frequencies, since it more correctly handles the motion of the air bubble as it moves away from the gun.

In this paper we extend our approach to obtain shot-by-shot estimates of the array depth and sea-surface reflectivity as part of the signature estimation. Our method follows from the somewhat unexpected observation that the near-field data can contain a strong ghost component, such that there is a noticeable sensitivity of derived signatures to the depth and reflectivity parameters used for their computation.

We use the estimated array depth and reflectivity values to construct directional far-field signatures containing a shot-varying estimate of the sea-surface ghost, and demonstrate shot-by-shot source de-ghosting via their use for directional designature. Our results show that, compared to modelled signatures, our near-field derived signatures contain a more accurate representation of the bubble and ghost components of the signature, leading to an improved treatment of wavelet phase and amplitude at low frequencies.