A stratigraphic pick in a well is a point defined in three dimensions (X, Y and Z). The accuracy of the pick, either in measured depth from the kelly bushing (KB), or with respect to sea level, is difficult to quantify. Sources of error include:
- incorrectly recorded well surface or bottom hole location, which results in an error in either X, Y, or both.
- incorrectly recorded ground or KB elevation, which results in an error in the Z dimension.
- incorrect stratigraphic top pick elevation values, or an incorrectly identified formation result in error in the Z dimension.
- errors in data entry or data transfer, resulting in errors in X, Y and/or Z values.
Any errors in the initial data points were reduced in the final data set by geostatistical evaluation to identify and remove outliers.
The vertical positional error in the sub-Cretaceous unconformity picks is unknown. However, there is a potential for the introduction of error due to inaccurate kelly bushing (KB) elevations, and mistakes made during selection of the stratigraphic pick.
Potential errors in KB elevation are likely the greatest source of vertical error in this dataset. The KB values are derived by adding the height of the drilling platform to the surveyed ground elevation. If this is not done correctly, it can introduce vertical error in the KB elevation, which is then propagated into error in the measured depth to the pick and therefore also impacts sub-sea pick depth.
Uncertainty in the vertical positional accuracy can also be caused by human error resulting from incorrect placement of the pick on the well logs.
The occurrence and magnitude of these errors are difficult to identify, however these types of errors were mitigated through geostatistical evaluation to identify potential outliers.
Step 1: 4023 stratigraphic picks of the sub-Cretaceous unconformity were produced for the study area.
Step 2: (geostatistical filtering): The points were geostatistically analyzed for the presence of outliers, defined as points with an elevation value that differs more than 20 m from the predicted value of the modelled surface. All outliers were reviewed, and adjustments were made where necessary to ensure all 4023 picks represented the sub-Cretaceous unconformity as accurately as possible.