by Tim Craske, Director and Principal Geoscience Consultant at Geowisdom
I have been modeling ore systems and investigating how ore bodies form and what makes them economic for nearly 30 years. I have applied my insights to find ore bodies and to help younger generations of geologists also do the same.
It has not been an easy or straightforward journey… and that’s the point. Any truly worthwhile and motivating goal must have a degree of difficulty, controversy and, if you are lucky, some serendipity too.
I started my targeting and project generation journey in the mid- to late-1980s when I moved from Western Australia to Townsville with Western Mining Corporation (WMC) to explore the then less-than-fashionable Eastern Succession at Mount Isa.
I nearly failed and lost faith in my mission. I came close to resigning from WMC to become a scuba diving instructor! What saved me was the support of a mentor: Douglas Haynes of Olympic Dam fame.
He was neither my first nor my last mentor. My first had been Professor Janet Watson during my studies at RSM Imperial College. My co-mentor at that time was a young but charismatic structural geologist called Rick Sibson. But it was Douglas who identified and nurtured potential at WMC in the crucial years of my early career.
Douglas taught what he called ‘holistic modeling’, also referred to as ‘conceptual modeling’. This was a combination of empirical observation, process science, a bit of thermodynamics and minerals chemistry wrapped up in critical thinking and the scientific method. It was an intoxicating mix for a young geo from the Cotswolds.
The next four years were spent in the Queensland outback, trying to explore the Eastern Succession outcrop and finding out it was even harder to identify original rock types there (due to extreme metasomatism) than it was to see through laterite in WA! I started to feel like a failure, which was tough. I lost faith and lost my way.
I was saved by the community of WMC’s Exploration Division and a geophysicist called Mike Webb. Mike and I formed a partnership of equals. Douglas still supported me in the background, but Mike was far more pragmatic: he suggested we explore for a target type we had evidence exists, rather than exploring conceptual Homestake-style sediment-hosted gold systems that we had already shown over four years did not exist in the Eastern Succession.
While Mike took an empirical approach, pointing out the extreme mag anomalies that might be Starra or Osborne look-alikes under shallow cover north of Cloncurry, I followed Douglas’ methodology and developed my own ‘Starra-style Au-Cu’ conceptual model. As luck would have it, our targets pretty much coincided. Both were driven by strong magnetics, but I favored complex ‘folded?’ structural signatures over extreme nanoteslas. The top 12 targets were nearly the same, and the middle strength response with greatest complexity was target number five… FC5, which became Ernest Henry. The number one target on the 100% WMC-owned ground to the east was E1. Both E1 and Ernest Henry have been mined profitably by Glencore’s Mount Isa Mines. Ernest Henry is currently being mined by Evolution and has been more profitable than Olympic Dam for most of the last 30 years. Not bad for a Tier Two IOCG that just happens to be close to Mt Isa’s smelter.
Conceptual modeling requires science, good thinking skills, and a lot of reading… interspersed with periods relaxing in nature, to allow the subconscious to offer up new insights. Like when snorkeling on the Barrier Reef or venturing into the Queensland rainforest, where Douglas encouraged me to experience what a stinging tree felt like! These were formative years; I was glad that I had not thrown it all in for scuba diving as a career and rock collecting as a hobby.
I have had several more mentors since leaving WMC. Among them are a couple of significant non-geologists, including an insurance salesman turned corporate coach and a car salesman turned hypnotherapist. My current mentors are both geoscience professors; I have other mentors on my personal board of advisers too, including management consultants and young AI gurus.
Back to the title of this article. What is an exploration model, and what is it worth? If you explore under cover without having a model, you have no hope of finding anything other than disappointment. But an exploration model is not a conceptual model. The latter incorporates sneaky ideas of how type deposits form. The former has to be expressed in proxy data that you can go out and collect to build and rate your targets. The conversion of the concepts into proxies is the key stage.
When you explore with an exploration model, you are not only testing the ground but also the model. If the model says good and yet you find nothing, the model must be questioned too. This is the scientific principle applied to the science of exploration and discovery. This is where critical thinking and systems thinking skills can help you, where candor and reflective thinking will make you and your model stronger.
So, what is a model worth? A lot… until its first failure.
As Einstein said, ‘If the model no longer works, change the model.’ Hunt for a better model as well as a better orebody.
‘We over-rationalize the past, we dramatize the present, and because of this, we underestimate the future.’
This is a set of remarks the President of Finland, Alexander Stubb, made at the Eurelectric Power Summit in 2026 that has since been widely shared by commentators and policy experts.
Given my work in systems thinking and mineral discovery, the quote aligns closely with several ideas this article investigated:
- Complex systems often produce non-linear change, making the future look more surprising than it actually is. This includes the probability of different outcomes from a single probability-based exploration prospectivity map or target-ranking schemes. These are either oversimplified or overly complex and can sometimes be hard to apply, except by experts.
- Exploration teams can become trapped by historical models, by ‘how deposits have always been found’ and also by today’s operational pressures, causing them to underestimate emerging technologies such as AI, new sensing methods, or paradigm shifts in ore targeting, including systems thinking approaches.
- It echoes the distinction between forecasting and foresight: rather than predicting one future or locking-in a single preferred or fashionable model, systems thinkers instead explore multiple plausible futures and models and prepare for and assess against their actual outcomes.
In the overlapping worlds of process, holistic, conceptual, empirical, and exploration models, one thing is clear, ‘All models are wrong… but some are useful.’ (quote from the British statistician George E. P. Box).
Box’s point was not that models are bad, but that every model is a simplification of reality. Since reality is too complex to capture perfectly, the value of a model lies in whether it helps us understand, predict, or make better decisions.
You make use of models at your peril, but this is still better than having no model; without it, failure results cannot be integrated into insights (transferable learning) and exploration strategies are hard to improve.
For more information: Get in touch with Tim on LinkedIn
