Seismic interpretation is crucial to mapping subsurface geological structures, understanding petroleum systems, and identifying geological risk associated with recovery.
Since the late 1940’s, Geoscience Australia has used seismic data to map the subsurface of Australia. As more data was acquired, data volumes grew. Adding to that, data acquisition technologies continued to advance, increasing data quality and generating larger data volumes.
To effectively interpret these data volumes, time often dictates the outcome of the interpretation. Having the time to look at every trace within a data set is practically impossible via traditional seismic interpretation methods. It is estimated that approximately 10 per cent of seismic data is interpreted by a Geoscientist within the time available. Added to that, in challenging market conditions, it is even more important operators continue to be as efficient as possible, without compromising safety. Decisions need to be taken with the best available information, whilst delivering the best subsurface understanding.
To allow operators to achieve this goal, Geoteric introduced an AI-backed seismic interpretation solution. Delivering clear improvements in subsurface understanding in a very short period of time, it is more than just a timesaving, data management platform. By nature, its data-driven design enables more coverage of the data than traditional methods, improving feature detection. Its true value; effective and detailed characterisation of geological elements which can enhance or alter an operator’s understanding of the subsurface, making or breaking the investment.
Since its release 12 months ago, Geoteric’s AI platform has had an immediate impact on operator subsurface understanding. Lead Geophysicist, working on the Valhall with AkerBP said; “this has improved the rigour of our work so that our decisions are based on a better understanding of where faults may introduce drilling risks, and so that we can better understand lateral changes in our reservoir”.
The quality of Geoteric’s AI-driven results are illustrated within the Canning Basin, Australia (see figure 1) which covers an area of 4,466km². In less than a day, faults of all scales were imaged. The level of detail exceeds what is possible with conventional interpretation, a process which can easily be months long. With the structure now revealed by the platform, the Geoscientist can effectively utilise their time to better understand the system.
Be it a high-level regional screening framework to determine where to focus efforts or at well resolution to highlight fault-related drilling risk, flexibility has been incorporated into the AI design. Combining both the strengths of human knowledge plus the capability of the machine, the platform is designed in such a way that results are easily integrated into established workflows for thorough quality control and ready for the next steps in the workflow.
Highlighting unseen detail at both a regional and local scale, results like this are unthinkable via traditional means. The platform has the ability to characterise features using all the data signals, ultimately improving feature detection confidence, all within a rapid timeframe. Geoscientists can interact with the results and by using Geoteric’s AI Seismic Interpretation, explore vast amounts of seismic data, derive insights and test assumptions to build a greater subsurface understanding.
To date, Geoteric have completed multiple commercial projects within Australia (including three for the same customer), and even more throughout the rest of the world to aid clients’ subsurface understanding and enhance field recovery. Revolutionising seismic interpretation, Geoteric have continued to develop their AI- driven seismic interpretation offering, delivering true seismic change.
Acknowledgements: Public data courtesy of Geoscience Australia.
