Accelerating Seismic Interpretation with Machine Learning [OLD]

Learn how Enthought can help accelerate your science

The schematic on the left shows an interpretation applied to an inline within the Poseidon seismic volume. The colored image demonstrates corresponding interpretation made by the automated system some distance from this initial interpretation. There is a high degree of consistency in what can be seen to be similar features.

In the second example, the LANDMASS-1 dataset was used. Example seismic facies were used to train the machine learning algorithm, seen in the tiles on the right, top to bottom; salt, chaotic, fault and horizon. The colored image is the automated interpretation of the F3 seismic data volume. The ability to deliver decision-ready quality image data with minimal expert time provides confidence in the potential of this new technique.



Basic stratigraphic units within seismic volumes are defined by careful analysis by a skilled sequence stratigrapher. These patterns may be defined objectively and are considered to be predictable.1 Partnering with a major independent, Enthought has conducted R&D into the efficacy of using machine learning techniques and custom processing workflows to automatically delineate these patterns throughout a seismic volume.

This initial seismic interpretation is time consuming for an expert, and requires rigorous attention to detail.




Machine learning and AI can alleviate the drudgery of interpreting large seismic volumes and allow more time for experts to focus on quality and value. Our approach is designed to operate where input from the interpreter is minimal. Where possible, we limit the complexity of machine learning systems, and apply standard seismic processing methods familiar to the interpreter, while taking significantly less time.




This approach is shown in the following two examples:

  1. The first is where a sequence stratigrapher provides a manual interpretation of an inline taken from the Poseidon seismic volume.2  Learning from the manual interpretation, a set of stratigraphic units are then automatically identified by the system some distance away from the initial interpretation.
  2. In the second example, the algorithm is trained using a library of seismic facies. This data was taken from the LANDMASS-1 dataset.3 The set of labeled seismic facies consists of salt, faulted and chaotic areas, and regular horizons. Learning from the LANDMASS-1 examples, facies are then predicted across the F3 seismic volume.4

Poseidon, LANDMASS-1, and F3 are all open datasets available to the public.


1Seismic Stratigraphy and Global Changes of Sea Level, Part 6: Stratigraphic Interpretation of Seismic Reflection Patterns in Depositional Sequences, R. M. Mitchum et al, AAPG Special Volumes, 1977

2The Poseidon seismic dataset

3LANDMASS, Large North-Sea dataset of migrated aggregated seismic structures

4F3, Netherland offshore F3 block complete

Learn more about how the combination of science and AI can help you