Comparison of Reflection-angle Estimation Methods – Accuracy and Sensitivity

Transforming a CDP-offset gather into a CDP-angle gather is one critical step for AVO analysis and angle-domain NMO de-stretch. The transformation from offset to angle usually assumes that (1) accurate velocity model; (2) the ray plane is perpendicular to surface; (3) the horizontal reflector; and (4) the hyperbolic normal moveout. Many studies have investigated the effect of assuming a flat model in the presence of dipping reflectors. In this study, we compare the accuracies of different reflection angle estimation methods, investigate their sensitivity to velocity perturbation within the overburden layers, and discuss the effect of a dipping ray plane. Simple modelling shows that reflection angle estimation is not sensitive to the velocity perturbation within the overburden layers above the target reflector, but very sensitive to the velocity errors near the target horizon. By using an optimized 6th-order NMO equation, the estimated reflection angle is accurate enough up to 60 degrees. 2nd-order NMO equation will overestimate the reflection angle substantially for larger reflection angle (>35 degree). The dip of the ray plane should be honoured when source-receiver direction deviates from the dip direction, otherwise, the reflection angle will be overestimated.