# Depth-velocity model building

Estimation of depth and velocity parameters is one of the main tasks of reflection seismic surveys. The process of velocity and depths estimation from wave field is called inverse kinematical problem solving.

Creation of a depth velocity model in Prime is layer-by-layer procedure starting from the upper layer and finishing with the lower layer. Travel times of reflected waves are estimated for current layer via velocity analysis in time or depth domain. These travel times are kinematically recomputed to the top of specified layer by ray tracing through all upper layers. Then interval velocity and depth are estimated.

In this way we always make computations as if we operate with homogeneous environment. In this case you will have a computational advantage e.g. you save time in interactive mode when changing current processing parameters and recomputing the results only for one layer and not for all of them. Processing can be extremely speeded-up in this way. Probability of mistakes collection is removed in Prime by correct way of computations.

Selection of number of layers for every specified time section is implemented with geological assumptions usage. The boundaries of interest, which are includes to the model must be related to the significant velocity contrast. Also they must be quite strong events with good coherency in velocity spectrum, computed along them.

Inverse kinematical problem is solved for current layer on the assumption about its homogeneity. Therefore it is needed to separate time section to the layers so as every of them will be characterized by constant interval velocity.

The way of data parameterization is choosing according to the geological assumptions (presence of absence of salt domes, reefs, gradients of velocity etc). The vital part of processing in Prime is the ability to verify your way of processing and therefore your results at every processing stage. To do that you can compute the special numerical criterion for every layer in your depth-velocity model. Such criterion is a powerful tool that helps you to avoid a lot of mistakes and uncertainties during processing.