Presenting automatic velocity model updating by reducing residual depth move-out in the presence of lateral velocity changes and velocity anomaly

Seismic velocity analysis is the main part of seismic imaging. Quality of the final seismic image, in velocity dependent imaging methods, strongly depends on the accuracy of the velocity model and consistency of the model with the data. Various types of methods and strategies were presented for velocity model building, such as gridded based and layered based velocity tomography inversion and migration velocity analysis. Conventionally, most of  velocity model building methods use an initial velocity model and some approaches to the final velocity model through several updating steps, mostly by means of the semblance values as the velocity picking criteria. However, this is a time consuming approach and mostly unreliable in the presence of strong velocity variations. Therefore, a new strategy was presented by introducing a parameter called  which is the ratio of the true velocity and the initial velocity. In this strategy, updating the velocity model would be performed through optimization of this parameter. Afterwards, depth imaging would be performed by the new velocity model and accuracy of the velocity would be evaluated by analyzing of common image gathers. For better parameter selection, the concept of the residual depth move-out in the common depth gathers was introduced. It defines the value of deviation in depth of the imaged reflector compared to its true depth. This strategy did not handle the effect of velocity anomalies and had some deficiencies in handling lateral velocity changes. Here, we present a new strategy based on the method of the residual depth move-out in the common depth gathers for velocity model building and depth imaging through appropriate selection of parameter , automatic correction and velocity picking during residual depth move-out correction on common depth gathers. In the presented strategy, we first define a range of possible  parameter for an initial velocity model, which can be obtained by any conventional method. Then, by defining a rational increment in this predefined range for parameter , all of the possible velocity models would be obtained. Subsequently, seismic depth imaging would be performed using all the possible velocity models. Afterwards, common depth gathers will be analyzed automatically to define residual depth move-out. By defining a threshold in correcting of this residual depth move-out, the common depth gathers that exhibit higher move-out than the threshold value will be corrected automatically. Here, those common image gathers that were affected by the velocity anomaly, would be analyzed by selecting a new range and new increment of the parameter to handle the effects of velocity anomaly. This procedure would be iterated until finding a satisfying depth image through depth imaging. This strategy was applied on a synthetic data and two field land data examples with lateral velocity variations and a layer with very high velocity value as velocity anomaly. Results have shown that the presented strategy can be considered as an alternative to the conventional velocity analysis methods.