The people, science and technology behind discovery

Subscribe to the Magazine

Get new articles sent directly to
your mailbox.


Join the network

Twitter Updates Group Forum

Earth Explorer is an online source of news, expertise and applied knowledge for resource explorers and earth scientists. Sponsored by Geosoft.

Upcoming Events

WindEnergy Hamburg 2018

September 25 - 28, 2018

SEG 2018

October 14 - 19, 2018

News & Views

All News


Rapid 3D gravity inversions aid uranium explorers

by Dan Zlotnikov on September 19, 2018 APPLIED

[Click to enlarge]

A ground gravity grid of the Contact prospect, with the area modelled with VOXI Earth Modelling outlined in black. The black lines show the projected surface traces of the inclined holes drilled in the area.
Image credit: Roy et al., Can. J. Earth Sci. 54: 869882 (2017)

[Click to enlarge]

The unconstrained (a) and constrained (b) 3D inversion gravity models generated in Geosoft VOXI. Adding the constraints split the single orebody into three smaller ones. The anomaly identified as "4" did not match drill core data, and was likely due to incomplete gravity coverage at the boundary of the claim.
Image credit: Roy et al., Can. J. Earth Sci. 54: 869882 (2017)

Mineral exploration must always contend with a degree of uncertainty: You never know exactly where the valuable deposit is, nor its precise shape, size, or the proportion of valuable minerals to host rock. Uncertainty is unlikely to ever entirely disappear from mining exploration, but technological improvements continue to help explorers manage it ever more effectively.

One notable application of technology was the Orano Canada’s use of the Geosoft VOXI Earth Modelling service in delineating a new uranium discovery. The exploration work took place in 2014 and 2015 at the Kiggavik uranium project in Nunavut, with the goal of outlining the newly discovered Contact prospect. The results are described in a paper published by NRC Research Press in May 2017.

Prior to Orano Canada’s predecessor company acquiring Kiggavik in 1993, the previous owner had already identified five deposits. Exploration continued at the site following the acquisition, until a price slump saw the project put on care & maintenance in 1998. Work resumed once prices recovered in 2006, culminating in the Contact prospect discovery in 2014.

While the discovery of another potential deposit at what was already a high-value project was big news for Orano Canada, the Contact prospect was also the first chance for the company to produce a 3D model of the alteration body from available gravity data. Régis Roy, geophysicist at Orano Canada and lead author on the paper, explains that previous targeting was done on a plan view – much like looking at a map from above. The geophysicists used a combination of gravity, magnetics, and very low frequency electromagnetic (VLF-EM) data layers to outline the location of the prospect. The geologists would then use the 2D results to position the drill holes. As each drill hole costs tens of thousands of dollars, having the best data available helps to define exploration targets before drilling.

However, a 2D projection is inevitably a simplified depiction of the three-dimensional target. Remote sensing techniques cannot reliably distinguish between a small deposit near the surface and a large deposit that’s located deeper down. To address this, Roy says, the team produced 2D gravity models of vertical cross-sections at numerous locations of the prospect.

“If the geological response we model in 2D matches the measured response in the ground, that means we're not too far from reality. Then the geologist will position the drill hole based on the geological hypothesis of this cross-section. The problem is, you have to do several iterations based on 2D sections, but you can miss,” he explains, adding that gravity line spacing at Contact was 100-200 metres – plenty of room to overlook a large, valuable target. In contrast, VOXI produces a 3D model of entire the target area, Roy says, “so you’re taking into account all the data in 3D.”

The Orano Canada geophysicists first produced an unconstrained 3D model of the initial gravity signature data in the area of Contact prior to the start of the drilling program. This model described as a large bean-shaped alteration anomaly, roughly 600m long and 200m wide and buried 120m below the surface. But while a major improvement over the 2D maps and cross-sections, this initial model was only the first step.

A brief explanation of Kiggavik’s geology may help understand the challenges faced by the explorers. The project is broadly similar to other basement hosted uranium deposits in Canada: The presence of uranium is hinted at by anomalies caused by hydrothermal alteration. The anomalies are formed when minerals deep underground dissolve in heated fluids. The mineralized fluids then travel until a trap involving a secondary fluid is encountered, causing uranium to precipitate in “alteration haloes” that are rich in clays, electrically conductive and non-magnetic—rendering them detectible from the surface through a number of standard geophysical techniques. The altered rocks are now less dense than the surrounding material, making them anomalies in gravity surveys. The catch is, an alteration halo may mark an area where uranium-rich fluids merely passed through without the valuable mineral precipitating As Roy puts it, “you can have good technical success, without finding what you were really looking for.”

Because the geoscientists could not yet tell which anomalies contained uranium, the unconstrained inversion was not sufficient in itself—it had to be shown to match the direct observations and physical properties measurements taken on drill cores.

Armed with the unconstrained inversion, Roy says, the team used the initial bean-shaped anomaly to position the first holes.

“Then they started drilling, and after 1-2 holes, we had physical properties to work with.”

Measuring the density of rock lithologies in the drill cores allowed the geophysicists to reduce the number of assumptions VOXI was making to fill gaps in data. The drill core measurements were providing the constraints for the 3D inversion, forcing a shape more closely reflecting reality underground. Each successive hole allowed more drill core studies to further reduced uncertainty. This, in turn, allowed the geophysicists to give the geologists better guidance on where to position the next drill holes. The process continued through successive iterations, Roy explains.

“We ran a constrained inversion to see how the geometry of the altered body changed with the addition of new data. Then, the geologists would continue drilling and provide more data, so we would continue to run inversions. We progressively use more physical data to constrain models until we got at final image of the altered body that resembles drill core results.”

Once the inversion accurately predicted drill core measurements, the explorers were done.

“At the end, we knew there was nothing more to explore in that area. At Contact all the holes are spaced 50m or less apart. And at the end, we knew we completely circled the gravity anomaly,” says Roy.

The final constrained model differed from the unconstrained inversion in some very important ways: The large bean-shaped anomaly was split into three separate orebodies within the volume of the "bean," closer to the surface than the unconstrained model predicted. The published paper suggests that in the final constrained model, “the location and the geometry of the low-density bodies are more realistic and validated by a geological model built from logging data.”

The paper further suggests that constrained gravity inversions such as the one produced at the Contact prospect can be useful in defining a drilling grid size for resource estimation.

It’s important to note that the ability to run 3D inversions is not sufficient in itself: The process involves vast amounts of data, is very computationally intensive, and can be very slow without powerful hardware. Because VOXI Earth Modelling is a cloud-based service, the technology is available at the exploration camp, only requiring an internet connection.

“After completion of the first holes at Contact, I was able to run four inversions in one afternoon by testing different parameters and providing a result to the geologist to reposition the adjacent hole the next day. Computational responsiveness and the ability to provide decision support were important aspects for this project,” Roy says.

Despite the promising results reported in the paper, exploration activity was halted at Kiggavik in 2016. A look at the uranium price makes clear why: Peaking at an incredible US$143/lb in 2007, uranium has remained below $45/lb since late 2012, currently trading at US$26.10/lb. Yet despite today's low prices, projections for future demand are strong. China, already the third largest consumer of uranium in 2017, had 38 operating reactors in April of 2018. But this number is set to almost triple, with 20 reactors already under construction and a further 41 planned to come online by 2030, taking installed capacity from today's 34.6GW up to as much as 150GW. Nuclear fuel consumption is expected to increase elsewhere in the world as well, if less drastically than in China: The World Nuclear Association reference scenario projects a 22% increase in demand from 2020 to 2030.

But the promising results produced at Contact won't stand still for a decade, waiting for prices to recover. Orano Canada has shifted its exploration focus to the Athabasca Basin in Saskatchewan, and Roy says his team is intending to continue applying and further refining the technique first deployed at Kiggavik.

“We aim to use and improve our methodology by integrating other geophysical data such as magnetics and EM data with the gravity data. We will also try to constrain the inversion with 3D geological models derived from drilling data. The use of VOXI Earth Modelling within the Orano geophysics team has a bright future ahead,” he predicts.

Roy also feels that the shortened feedback cycle can offer benefits in a wide range of exploration contexts.

“Most ore deposits are accompanied by hydrothermal alterations that can cause strong changes in the rock’s physical properties leading to significant geophysical anomalies. Density and magnetic susceptibility, for example, can increase or decrease depending on the nature of the host rock and alteration processes. The inversion of any potential methods with VOXI Earth Modelling can definitely help geologists who are exploring other commodities like iron, nickel, copper and gold to have a better idea of the geometry of the alterations haloes that characterize their deposits.”


View all articles »