Labeling Data during Horizon Gridding

May 1999

When using horizon gridding, separate grids for each fault block are created from a single data set. It is critical, however, to be sure that each data point is used in the correct fault block for gridding, otherwise a data point can incorrectly influence the shape of the horizon.

The EarthVision 5 release contains a function for labeling each input scattered data location with the name of the fault block in which the data point was used for calculating a horizon. By viewing the labeled data in the 3D Viewer, users can get an immediate visual understanding of how the data points were used in horizon gridding, and if the data points were used correctly. This example illustrates how to use the labeling function contained within the EarthVision 5 horizon gridder.

The following image shows a portion of a faces file with three seismic lines of data posted.

Data from three seismic lines posted on top of a four-layer model.

This particular view shows the completed horizons; however, typically users would first look at the data in relationship to just the fault surfaces.

In the first step of horizon gridding, the data are labeled based on which fault block they lie in. In this particular data set, some problems appear in the line farthest from our eye. The figure below shows the same seismic data color-coded based on which fault block it was used in horizon gridding. (For information on how to create the labeled data set in the WorkFlow Manager and/or the Geologic Structure Builder, refer to Labeling Horizon Data in the WorkFlow Manager and Labeling Horizon Data in the Geologic Structure Builder, respectively, below).

Seismic data color-coded based on for which fault block each data point was used in horizon gridding. The marked point is being used for the wrong fault block.

The point at the 3D Cursor location is colored in yellow, indicating that it is associated with the fault block on the foot wall of this large reverse fault (the data point is being used to model the horizon in the foot wall because of its lateral position relative to the fault). Clearly, based on its elevation, it should be associated with the fault block of the hanging wall. If modeling proceeded with the data being used as labeled, the horizon in the foot wall would be pulled upward to try to honor the mislabeled point.

Clearly, an acceptable horizon will not be calculated with this data set in its current form since several of the points are labeled in the wrong fault block. There are two solutions to this problem. The best solution is to correct the fault surface so that all the points that should be used in the hanging-wall fault block would be physically within that block (the fault-surface location should be changed usually because typically the accuracy of the fault location is less than the accuracy of the data location).

A second solution, shown here for illustration purposes, is to increase the "fault tolerance" value. The fault tolerance is used to filter out points within a certain distance of a fault surface. These filtered points appear as white in this following illustration.

Seismic data color-coded based on for which for which fault block each data point was used in horizon gridding. The white/gray points are ones not used in the horizon gridding calculations due to an increased fault tolerance value.

This image shows exactly which points are not used in the horizon gridding process when a fault tolerance is applied.

A more extensive discussion of horizon gridding techniques in EarthVision 5 is found in the FAQ section of this web site and, of course, in the EarthVision User's Guide.

Labeling Horizon Data in the WorkFlow Manager

Labeling data can be performed in several ways in the WorkFlow Manager. It is recommended that the data be labeled during the first pass at horizon gridding to save time. The easiest way to label data is to check the Validate After Computation box for the Horizon Modeling step. During the Horizon Modeling calculation, the color-coded, labeled data is then displayed in the 3D Viewer (without the fault surfaces). The user is given the option of changing or correcting the data and then recalculating the horizon on the fly in the 3D Viewer, or canceling the Horizon Modeling step. An alternative method is to select the Compute Fault Preview Faces File and Save Labeled Fault Data check boxes on the Fault Data Parameters window or the Compute Fault Faces File After Fault Tree Building and Save Labeled Fault Data check boxes on the Fault Tree Building window. In either of these cases, a 3D Viewer icon becomes available next to the Fault Modeling and Fault Tree Building steps (respectively) in the Computation window. Selecting the 3D Viewer icon brings up the fault faces file along with the labeled, color-coded scattered data.

Labeling Horizon Data in the Geologic Structure Builder

In the Geologic Structure Builder is the option Calculate -> Horizon Grids -> Label horizon data (on the main menu bar). When this option is selected, the user is prompted for a directory name and then the labeling is performed. When the labeling calculation is run, a data file for each of the horizons is written into this directory. The data can be viewed by itself (i.e., without the fault or horizon surfaces) by selecting the View push button on the Job Done window. The data are displayed in "zone" colors (where each zone color is related to a fault block) so one can easily see the segregation of the data by fault block. If, however, this data set is displayed with a faces file, the data are automatically displayed in property colors. To turn on zone colors, after loading the faces file and the data file (displayed via the Edit Data menu), on the Color menu, select Scattered Data Colors button and then choose Zone Colors.