Is grade distribution in mineral deposits controlled by lithology or structure?
Resource geologists around the world talk about ‘lithology models’, ‘grade models’, and, on rare occasions ‘structural models’ as if they were completely separate and unrelated features that can be modelled from drilling data. This article discusses this issue—I hope to convince you that this misrepresents reality.
Most geologists mean ‘fault model’ (a model that displays the position of faults in space) when they use the term ‘structural model’, irrespective of whether the bulk of the deformation is brittle or ductile. Resource geologists apparently consider faults to be most significant structural features so ‘fault models’ have become synonymous with ‘structural models’ over the last several decades. In fact, I have found that the word ‘structure’ is virtually synonymous with ‘faults’ in the mining industry. Very few geologists use the word ‘structure’ when they are referring to distortion effects caused by ductile deformation. Over time, these resource professionals view models of grade, lithology, and structure as separate and discrete entities, presumably because they’re created using separate tables of drilling data (Figure 1).
Debate then surrounds ‘grade’ distribution and whether it is controlled by ‘lithology’ or ‘structure’ (Figure 2). One of these is usually thought to control the grade distribution, but which is more important at a given deposit—structure or lithology? It would be unfortunate, and perhaps calamitous, if we studied the lithological distribution if it has no bearing on the grade distribution.
I’m a structural geologist who makes a living by studying 3D grade distributions, but this debate about grade distribution being a function of lithological control or structural control makes no sense whatsoever. It’s as if resource geologists have made up their mind that lithological distribution, in some deposits, has nothing to do with deformation. This artificial and illogical separation of lithology and structural geology is an unwritten, but apparently a universally accepted, truth among resource geologists, and seems to have passed down from one generation to the next like well-loved folklore.
A client recently told me that the lithological model I delivered to them was ‘simply a lithological model and not a structural model’ that they had asked for. While they are entitled to their view, such a position is utterly illogical to me. Let me explain what the rocks looked like at this deposit so you can get an idea of the context here.
These sedimentary and granitic rocks had undergone intense and pervasive schistosity development under greenschist facies metamorphism and the foliation is now sub-parallel to the major lithological contacts. There is also evidence for isoclinal folding and fold limb attenuation that resulted in rootless folds. The lithological units are now dipping at a steep angle of 70–80 degrees. There are some minor faults, but these had little effect compared to the pervasive ductile strain that affected 100% of the rock units and controlled the grade distribution, which was emplaced during the ductile deformation.
To a structural geologist, talking about the host rocks as if the lithological distribution is completely independent of structural architecture makes no sense in the context of this level of ductile deformation. This point can be illustrated by viewing a few images of grade and lithological distributions from similar deposits that underwent deformation under similar conditions.
Figure 3 shows a down-plunge view of four naturally occurring grade distributions of Ni, Au, Cu, and Ag (not necessarily in that order). In all cases, the high grades mimic the synformal architecture of the lithologies.
If we model the grade distribution and create a wireframe outline of the high grade zones and define them as volumes, would the generated 3D model be considered a grade model, a structural model, or both?
Surely you can see that the grade is a function of structural architecture, so a model of grade would be both a grade and also a structural model, right? No, not according to conventional wisdom where only brittle faulting strain is defined as being ‘structural’. The grade distribution is controlled by pervasive ductile strain, not by brittle fault strain, so these models would not be traditionally regarded as a ‘structural model’ as defined by resource geologists; instead, they are regarded as ‘grade envelope models’ with no structural information.
However, I would argue that these grade models are still legitimate structural models
If a ‘structural model’ was built under the guidance of the resource geologist (i.e. fault model), then we still wouldn’t know what the mineralisation controls are because these deposits are all controlled by pervasive ductile strain, not localised faulting. This creates a strange situation where the answer to the problem of structural control of the mineral deposit is staring right at us in the form of grade distribution, yet we can’t recognise its significance!
Using a similar argument to the relationship between grade and structural geology (above), Figure 4 shows two modelled lithologies that are folded. A pervasive axial plane cleavage is present in both lithologies. So, is this model a lithology model, a structural model, or both? Again, it is clear that the lithological distribution is the function of pervasive ductile strain, therefore the model is both lithological and structural.
Figure 5 represents the overall relationship between grade, lithology, and structural geology that I’ve observed from hundreds of mineral deposits—grade and lithological geometries are a subset of structural geology, therefore grade and lithology are inseparable from deformation. Modelling the lithological distribution or grade distribution will tell you a lot about the structural nature and control of these deposits.
This is the approach I took in my latest paper on the Sigma-Lamaque gold deposit from Quebec (Cowan 2020), which you can download from here for free. In this case, the gold mineralisation at Sigma-Lamaque is very late and brittle in character, but the gold distribution is controlled by earlier formed structural architecture formed from folding, which in turn controlled the geometry of the post-fold diorite intrusions which were later fractured with gold-bearing veins. I identified a nested system of structural control which I describe in this article on LinkedIn using an understandable analogy.
The major control of lithological and grade distributions in nearly all mineral deposits is structural geology, because strain results in structural permeability. This permeability allows mineral-bearing fluids to flow through otherwise less permeable rock masses. In some cases, the correspondence of grade along a particular lithology (explaining the overlap of the two fields in Figure 5) can be demonstrated, but usually this is because the lithology had some properties that favoured mineralisation (commonly permeability), so ultimately structural geology controls the mineralisation.
As stated above, geologists can usually identify faults and narrow shears in core and in outcrop, but they often fail to recognise pervasive schistosity and its significance on the control of ore deposits. This results in them believing that the grade distribution is primarily controlled by lithology and thus results in misunderstanding the control of ore emplacement. This scenario is illustrated in Figure 6 where the influence of grade distribution is understood to be largely lithological with primary geometries, with little or no evidence for structural control. Where structural control exists, it’s almost always interpreted to be brittle faulting controlling the mineralisation. I would argue that this scenario is unlikely to be common in orogenic settings, except in porphyry deposits, which are largely brittle and low-strain. I intend to discuss specific cases in detail in peer-reviewed articles.
Another point worth re-emphasising is that many mineral deposits are the result of brittle-ductile deformation (e.g. orogenic gold), or brittle deformation that has affected a large volume of rock mass (e.g. porphyry style mineralisation). These mineralisation styles affect a large volume of rock mass—mapping one or two isolated faults seen in drill core won’t resolve anything with certainty. In most situations, joining a single fault seen in one drill core to another fault seen in an adjacent core is impossible. Such exercises of ‘modelling’ faults or veins are futile and are a great example of ‘busy work’ that keeps people employed, but resolve nothing of importance. Instead of wasting time interpreting faults, it is far more instructive to look for the ductile features that are hiding in plain sight, or identify large-scale structural features that control the broad grade distributions at the deposit-scale, then place your core observations in the context of such large-scale understanding. Modelling of the lithological distribution then becomes just one part of the structural analysis process.
Looking at the entire drilling data in 3D at the deposit-scale gives you valuable structural information and helps you understand lithological distribution—you can’t get that from measuring structural data from core in isolation. If we don’t understand the large-scale context, there is little chance we can understand the finer details we see in core or in outcrop, and that’s one of the main messages of my recent paper (Cowan, 2020).
References
Cowan, E.J. (2020) Deposit-scale structural architecture of the Sigma-Lamaque gold deposit, Canada—insights from a newly proposed 3D method for assessing structural controls from drill hole data. Mineralium Deposita 55, 217–240 [https://link.springer.com/article/10.1007/s00126-019-00949-6].
