A combination of geophysical techniques, data and sensing is taking the uncertainty out of minerals exploration, allowing companies to focus their efforts on where it counts early on and hit targets with greater precision. TIM THWAITES reports

Article from resourceful: Issue 9, March 2016



Seeking exploitable ore deposits beneath the Earth’s surface has many features in common with an x-ray of the human body.

In both cases, a volume is scanned to detect and interpret changes in the measurements of physical properties. And equally, the search is for something out of the ordinary – an anomaly.

But, there is a critical difference. When scanning the human body, we already know its boundaries, where we ought to find the bones, organs and tissues, and how they should appear.

Whereas under the Earth's surface the territory is boundless and we typically only have a vague idea of where the anomalies – ideally mineral deposits – should occur and what form they might take.

CSIRO researchers are working with mining companies and other government agencies on ways to increase the effectiveness of geophysical techniques, which can be used to explore for minerals under cover.

Mineral deposits are typically dense, often magnetic and usually metallic. That means they’re likely to cause a distinct change in the gradients of magnetic or gravity fields or an ability to conduct electricity.

Gravity, magnetic and electromagnetic surveys can often be used to detect orebodies under cover, in combination with other geophysical techniques, data from geochemistry and remote sensing using light and radio waves.

Dr James Austin, a CSIRO geoscientist, has been using magnetics data to assist mining exploration in northern and central Australia. One of the useful outcomes of the work is the ability to identify poor exploration targets early, allowing more funds to be spent on better prospects.

In one case, a company working in the Arunta block in Central Australia had magnetic data that suggested potential nickel and copper resources.

"We measured the magnetic properties of the rocks in the area and they all corresponded to a specific geological event," Dr Austin says.

"This allowed us to model the potential deposits accurately. Once we knew their volume and how close to the surface they were, we could estimate how expensive they would be to mine.

"If we assumed a conservative percentage of mineralisation, we could also estimate how much they were potentially worth."

The company was immediately able to dismiss 90 per cent of the potential orebodies as commercially unviable and concentrate on the ones that showed the greatest promise for development.

To take a step back to the human body analogy, we know where things should be, so we can model the information we would typically expect to find in our measurements and then use that model to detect and interpret anomalies.

With geophysics, the opposite is true. It becomes an inverse problem. We only have a vague idea as to what to expect, so we attempt to determine the nature of the orebody from our actual measurements.

Unfortunately, there is an infinite number of models of different rock types and orebodies at different depths that can fit a particular set of measurements.

It's possible to eliminate many of these models, because they don't make geological sense. And the more information that's accumulated about the area of interest (such as geology, other geophysical and chemical measurements, and the history of its development) the more the models can be refined to explain the measurements.

Mining exploration companies want accurate answers rapidly and at a reasonable cost.

CSIRO geophysicist Dr Juerg Hauser is working with industry to give answers using Bayesian approaches, which are updatable and provide a likelihood as to how correct they are.

"Given an unlimited amount of time and an infinite amount of computational resources, you could try all the models you could think of," Dr Hauser says.

"Or, you could cut costs by using an efficient deterministic inversion that quickly finds you one single model, but by doing this you would learn little about equally plausible alternative models.

"What I'm trying to find are pragmatic compromises – techniques that are fast enough, but comprehensive enough so that we have an understanding of the uncertainty – the range of models that fit the measurements made in the field."

Leader of CSIRO's exploration through cover research group, Dr Tim Munday, has been working in parallel, applying

Dr Hauser's and other software to airborne electromagnetic data.

Dr Munday says this data is good for targeting some ores directly, but also for gaining a picture of the regolith cover.

"The regolith information can also be used to trace ancient water courses, which can indirectly point to orebodies, particularly uranium," he says.

resourceful magazine

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.

Contact us

Your contact details

First name must be filled in

We'll need to know what you want to contact us about so we can give you an answer.