Researchers from CSIRO’s mineral resources business and digital innovation company Data61 – the largest data innovation group in Australia – have been looking at ways of sifting through and adding value to the vast amount of data gathered and held by Australian and state-based government geoscience agencies.
Article from resourceful: Issue 9, March 2016
As first cab off the rank, they have developed cloud-based machine-learning techniques that integrate different types of geological data to generate maps that show potential for successfully prospecting for nominated minerals across the entire continent of Australia.
According to CSIRO's Dr Jesse Robertson, the software is able to learn from geophysical data and information from other sources such as geochemistry, mine databases, remote sensing and geological mapping. And, it can continue to learn and improve as new data flows in.
"It basically gives you a prospectivity map: a probability that an orebody is sitting at a particular location plus how certain we are of that estimate," Dr Robertson says.
The uncertainty measure incorporates the quantity and quality of the data on which the estimate is made and can suggest where more exploration work should be focused.
"It's another tool in the tool box, flagging areas as interesting or not," Dr Robertson says.
"If an area is interesting, you could go and do a proper inversion of geophysical data or collect more data."
The new tool is yet to be publicly released.
resourceful brings you the latest in innovation, technology and research from across the minerals value chain — from mineral exploration to mining and metal production. Get it by email or in print.