The focusing process for time imaging is improved drastically when high-density parameter fields are used. Large offsets, steep dips and finally the anisotropy of the subsurface revise the bases of time processing. Today, two parameters are required: velocity (V) and anellipticity (η). Picking V and η using two-pass techniques cannot be a long-term solution. The estimation of both parameters is very sensitive to the mute function separating near to far offsets. Picking both parameters simultaneously using dense bispectral analysis overcomes this situation.We are proposing in this paper an original parameterization of the non-hyperbolic moveout, which increases the sensitivity of the analysis and allows static moveout corrections, necessary for automatic dense pickings.An intelligent QC sorting of the raw V and η fields, based on lateral coherency of the semblance and the Dixinversion ability of local V and η functions, prepares skeleton fields for simultaneous geostatistical filtering of both parameters.

No abstract

No abstract

No abstract

No abstract