LIDA Data Scientist Development Programme

Project Summary

The aims of this project are to build a state-of-the-art data mining tool to extract useful rock mechanics data from the vast legacy of published papers and open access theses; and secondly to use a statistically significant dataset to test the hypothesis that rock failure strength can be predicted using only the rock texture.

This innovative project will link collaborators in the School of Earth & Environment with those in LIDA and an expert partner, Dr Rob Cuss, in the British Geological Survey. The legacy dataset of rock mechanics test data goes back almost 100 years, but the relevant data are in unstructured text format – typically PDFs of published journal papers or open access Masters and PhD theses.

The combination of data mining and statistical analyses will provide a novel and useful tool for all agencies involved in the management of subsurface rocks and fluids. As the Energy Transition accelerates, the demand for useable geomechanical data – especially the strengths of different potential reservoirs, aquifers, or storage sites – will massively increase. We will test and refine a novel method based on open data that could make this process quicker and cheaper, producing a public good and thereby enabling more rapid decarbonisation towards net zero.

It is an exciting project that could make a huge impact in the geological community. I am honoured to contribute to this innovation.