Optimal processing of marine high-resolution seismic reflection (Chirp) data

Chirp frequency-modulated (FM) systems offer deterministic, repeatable sour ce-signatures for high-resolution, normal incidence marine seismic reflecti on data acquisition. An optimal processing sequence for uncorrelated Chirp data is presented to demonstrate the applicability of some conventional sei smic reflection algorithms to high-resolution data sets, and to emphasise t he importance of a known source-signature. An improvement of greater than 6 0dB in the signal-to-noise ratio is realised from correlating the FM reflec tion data with the transmitted pulse. Interpretability of ringy deconvolved data is enhanced by the calculation of instantaneous amplitudes. The signa l-to-noise ratio and lateral reflector continuity are both improved by the application of predictive filters whose effectiveness are aided by the repe atability of the Chirp source.