Other Methods

You can predict that the best log-editing method of the future will not be fully automated, because of the nature of the editing problem. The art of log restoration is the art of diagnosing which curves are valid at each depth. Log quality varies according to non-mathematical influences, such as:

• Logging Run Changes
• Borehole Rugosity
• Exotic Lithologies
• Whether the log was recorded on a major holiday
• etc

Here are some simple tests you can apply, to determine whether your log editing has been successful:

• Plot the amount of correction against depth, with logging run changes annotated … Is the data best at the bottom of each run?
• Check the integrated traveltime from the sonic log against the checkshot times, at every depth for several wells … is the correspondence consistent?
• Do synthetics tie perfectly? (they will be better than the raw logs, but are they as good as they could be?)
• Do the well-based wavelet extractions make sense spatially?