Real-Time Detection of Karstification Hazards While Drilling in Carbonates

The nature of carbonate deposition can cause the development of unique geological features such as cavities and vugs called karsts. Encountering karsts while drilling can lead to serious consequences. To improve drilling safety in intervals of karstification, it is important to detect karsts as early as possible. The use of state-of-the-art geophysical methods cannot guarantee early or even real-time detection of karsts or karstification zones. In this paper we demonstrate, based on an analysis of 20 wells drilled in karstified carbonates in the Barents Sea, that a karst that is dangerous for drilling is often surrounded by one or more other karstification objects, thus forming a karstification zone. These zones can be detected in real time through certain patterns in drillstring mechanics and mud flow measurements. They can serve as indicators of intervals with a high likelihood of encountering karsts. The identified patterns corresponding to various karstification objects are summarized in a table and can be used by drilling engineers. Apart from that, these patterns can also be utilized for training machine learning algorithms for the automatic detection of karstification zones.