f Accurate Simulation of the Kinematics of DMO in Travel-Time Inversion

In travel time inversion, times are picked in different domains, such as common shot, common mid-point or common offset, after some processing has been applied to clean up the data. One classical process is DMO which is nowadays routinely applied to seismic data sets. DMO modifies the location of reflections both in space and time. As a consequence, there are two solutions for a correct travel time inversion either reprocess the data without applying DMO, or, alternatively, take the DMO process into account in the inversion. The first solution is particularly time-consuming, especially when dealing with stacking velocity inversion in 3-D. In this case, in fact, post-DMO stacking velocity picks are a standard piece of information readily available from the seismic processors. This paper wilt therefore be dedicated to the second solution that consists of accurately simulating the kinematics of the DMO proces in the modelling phase of inversion.