Cratonic settings with mixed lithologies

In cratonic settings, you usually find a large range of geological formations with different lithologies, generating a pronounced velocity stratification that can only be captured in a layer-cake velocity model. Unconformities are common and may have locally removed some of the velocity layers. Often this stratigraphic record is overlaid by a Quarternary-Tertiary interval of compacting clastic sediments.

With these conditions, you will need to perform the following tasks:

  1. Decide on an appropriate velocity layer break-down.

    In general it is considered good practice to keep the number of layers small: many layers (> 5-7) means many potential sources of perturbations and frequent problems with unreliable velocities calculated for thin layers.

  2. Make all base surface time horizons (2D grids) defined at all nodes, including clipping for truncated horizons. Interpolate horizon inputs with a kriging or distance weighted option, or use an equivalent Property Calculator option (do not overwrite original horizons in 2D grid type, but generate copies for velocity modeling purposes). Note that the fidelity requirements for velocity horizons are much less strict than for structural horizons. Then clip 2D grids using the tool in the 2D Grid Tools > Removal group (prepare > Post-Processing Tools), or use a Property Calculator expression.
  3. Extract time values for well markers and calculate interval velocities. If the well markers and 2D grid time horizons have identical names, you can copy the information from the Marker table to a spreadsheet. Right click in the Marker table and select Copy with headers. Use the spreadsheet tools to analyze the well markers depth and horizons times. For individual markers you can also use the Report > Intersection Overview Report option from the context menu of the well marker in the JewelExplorer, or, to select all markers use multi select and then Report > Intersection Overview Report.
  4. Use regression techniques to analyze relation between Vinst and depth to be captured in k factor, eventually suppressing unreliable samples. In all cases, regressions should be optimized for minimum depth residuals, e.g., by using the Excel Solver Add-In. If you choose to work with laterally varying V0 values then export the V0 values per base surface together with the corresponding X, Y coordinates plus Z = 0 to a text file and import it as a point set (Data > Miscellaneous > Point Set), using the 'User-defined Formatting for Column Files' format.
  5. Interpolate V0 values to define velocity property everywhere on 2D time grids, e.g., by using a Property Calculator function as described above.
  6. Set up a velocity model.