The insights Enhanced oil recovery is a focus for oil companies, particularly in Norway where the government pushes them to demonstrate plans to recover at least 50 percent of the initial oil in place.
4D seismic surveys are critical to managing these high recovery rates. Repeatability—the similarity of seismic traces from the same location from one survey to the next—is as critical as quality of data within a survey. To date, the biggest breakthrough has been in the deployment and operational handling of permanent seismic arrays on the seafloor. With the receivers now in the same place each survey, the next area for improvement is the seismic source vessel which is subject to many variables including the weather and tides, as well as its speed, direction, and operation, as it sails the open seas..
To quantify and analyze the various behaviors of the source vessel for ‘The Sailor,’ data were integrated from a myriad of different sources including tidal models, weather records, GPS receivers on the vessel and array, gun depth and pressure sensors, and seismic processing attributes. The amount by which a seismic-source vessel crabs (heads into the wind to maintain its sail line) emerged as a key concept to represent environmental factors.
As expected, environmental conditions clustered shot points according to the sail line. Our visualization uses Teradata Aster® Analytics to represent these shot-point pairs. It shows how pairs cluster in similar conditions, and how the similarity of conditions affects repeatability. Three major populations of nodes (dots) represent the shot points. The right-hand limb refers to shots made while crabbing with a tailwind. The left-hand limb reflects crabbing with a headwind. And the upper limb shows calm sailing conditions. Redder nodes and edges (lines between nodes) reflect higher repeatability, beige moderate, and blue poor.
A better understanding of when sailing conditions are good or not-so-good (repeatability-wise) gives the source vessel a better idea of which line to sail in a given condition. After all, in heavy seas with a choice of lines 100 meters apart, which one would you pick? The one that’s planned but may have been conducted in heavy seas in the last survey? Or the adjacent line, acquired in calm seas in the survey before last?
Taking this a step further, processing techniques could be enhanced, using inputs from the sensor data to even out vessel-based differences and reduce survey-to-survey variations..