Seismic Data Engineering for Machine Learning Inversion

Kai Zhuang, Marcelo Guarido, Daniel O. Trad, David J. Emery

The CREWES Data Science Initiative presents the second lab of a series of four, focused on the solutions for seismic inversion using machine learning.

Geophysics in the Cloud was a competition with the goal to perform seismic inversion of rock atributes from seismic data with the use of well logs. It used open data (3D Poseidon from Australia) and the competitiors needed to perform inversions for P-Impedance, S-Impedance, and Density. Well logs with DTC, DTS and RHOB are used for training and evaluation (two blin wells). For the modelling part, the competition provide 4 different volumes: near, mid, and far stacks, as well a smooth P-velocity model from semblance analysis. Only these four input features may not be enough to get a reasonable inversion of the three target attributes.

Kai Zhuang is a PhD student at CREWES under the supervision of Dr. Daniel Trad and, for the competition, worked to create different seismic attributes from the near, mid, and far stacks to help in the inversion modelling. All in Python.

Kai will do a hands-on demonstration during the learning lab, and will use the same dataset provided in the competition.