Exploration companies are naturally drawn towards mineralisation.

We drill targets to find the "juicy" zones. Assays are tracked closely, laboratory QA/QC is scrutinised, and strong intersections quickly attract the attention of technical teams, management and the market.

That attention is justified.

Without representative sampling and reliable analytical results, there is no defensible Mineral Resource estimate.

But finding mineralisation is one milestone.

Having the information required to estimate it, classify it and demonstrate that it has reasonable prospects for eventual economic extraction is another.

The distinction often becomes painfully clear when a project moves from exploration into geological modelling and Mineral Resource estimation.

The question changes from:

Have we found mineralisation?

to:

Is the information supporting it actually ready to become a Mineral Resource?

I use "resource-ready" here as a practical description, not as a formal JORC term. By resource-ready, I mean an information base with sufficient integrity, spatial control, QA/QC, geological support, physical property data, traceability and supporting documentation for a Competent Person to assess whether a Mineral Resource estimate and classification can be defensibly supported.

And importantly, that assessment extends beyond the geological model itself.

QA/QC does not mean assay QA/QC

Mention QA/QC on an exploration project and the conversation often goes immediately to Certified Reference Materials, blanks, duplicates, laboratory performance and assay batches.

Those controls are essential.

But QA/QC should not be synonymous with assay QA/QC.

Every material dataset that eventually feeds into interpretation, modelling, estimation and classification should have an appropriate level of quality assurance, quality control and documented confidence.

That includes:

• collar locations;
• downhole surveys;
• geological and structural logging;
• recovery information;
• density measurements;
• geochemical tools such as pXRF;
• database capture and transformations; and
• the metadata describing how, when and with what equipment the information was collected.

JORC Table 1 reflects this much broader view. It asks about the accuracy and quality of collar and downhole surveys, drilling methods, recovery, logging, verification, data spacing, orientation, database integrity, bulk density, geological interpretation and classification, not simply sampling and assays (JORC, 2012).

The confidence in the final estimate is therefore accumulated across the entire data chain.

Two numbers in a database are not necessarily equal in confidence

Take collar coordinates.

Two sets of Easting, Northing and RL values may occupy exactly the same database columns.

But how were they obtained?

Was the collar located using a handheld GNSS? Was it subsequently surveyed using an appropriate survey-grade RTK system? What instrument was used? What controls were applied? What accuracy was expected? Was that information recorded?

The coordinate itself is only part of the data.

Its provenance and expected confidence matter as well.

That does not mean a handheld GPS measurement has no value. Its suitability depends on the stage of the project and the purpose for which it is being used.

But when those data begin informing a Mineral Resource, it becomes increasingly difficult to treat different levels of positional control as though they carry the same confidence.

JORC specifically asks for the accuracy and quality of the surveys used to locate drillholes, including both collars and downhole surveys (JORC, 2012).

The same thinking applies downhole. Modern survey workflows can now include QA/QC at or near the point of collection. For example, IMDEX HUB-IQ provides survey validation, planned-versus-actual comparisons, QA/QC reporting and approval workflows for downhole survey data. The particular product is not the point. The important development is that quality control is increasingly being applied to where the hole actually went, not merely what grade came back from the laboratory.

Density needs QA/QC and spatial representivity

Density is another obvious example.

Density is required to convert interpreted volume into tonnage. It therefore directly influences estimated tonnes and contained metal.

Yet Abzalov (2013) observed that density can receive substantially less attention than assay grades despite its importance to Mineral Resource estimation.

A density program needs more than a spreadsheet containing a large number of measurements.

It needs:

• a documented and appropriate measurement method;
• routine balance checks and calibration;
• traceable check weights across the operating range;
• repeat measurements;
• appropriate treatment of porous or weathered material;
• consideration of moisture and void space; and
• an understanding of which geological materials the measurements actually represent.

JORC specifically requires consideration of the frequency, nature, size and representativeness of density samples and differences between rock and alteration zones (JORC, 2012).

That last word, representativeness, matters.

Five hundred density measurements concentrated in one accessible part of a deposit do not necessarily provide better support than a smaller but properly distributed dataset.

You need a reasonable spread across the material you intend to estimate: across relevant lithologies, alteration and weathering domains, mineralised areas and, where material, vertically and laterally through the deposit.

The same philosophy applies to QC. If a balance is routinely operating across a broad mass range, known check weights, for example 0.5 kg, 1 kg, 2 kg and 3 kg where appropriate to the equipment, can provide routine evidence that the measurement system remains under control.

They are not CRMs in the assay sense.

But the principle is familiar: measure something known so that you can have confidence in what you are measuring that is unknown.

A clean logging database is not necessarily validated geology

Logging needs similar attention.

Database validation often asks whether FROM and TO intervals overlap, whether codes are valid and whether mandatory fields are populated.

Those checks are important.

But a perfectly clean database can still contain poor geology.

Logging should also be validated against the physical evidence.

Does the logged lithology make sense when checked against the core or RC chips? Are contacts and weathering boundaries consistent? Are alteration and mineralisation codes being applied consistently between geologists?

For diamond core, has appropriate basic geotechnical information been captured, such as core recovery and, where relevant, RQD? Has structural information been captured where it could materially influence later interpretation or mining assumptions?

JORC Table 1 specifically asks whether core and chip samples have been geologically and geotechnically logged to a level of detail appropriate to support Mineral Resource estimation, mining studies and metallurgical studies (JORC, 2012).

I am not suggesting every early exploration hole requires a full geotechnical program.

But information such as core recovery, basic RQD on diamond core and structural observations is relatively inexpensive to collect when the core is in front of you.

Trying to recreate it years later is another matter entirely.

Where pXRF or other supporting geochemical information is available, it may also provide a useful independent check on geological logging, for example, identifying changes in lithology or geochemical boundaries. That does not require pretending that field pXRF is equivalent to laboratory assay. Its own QA/QC needs to be fit for purpose. The point is to use independent evidence intelligently to test the interpretation rather than simply accepting that because a geological code exists in a database, it must be correct.

Resource-ready also means thinking beyond the geology

There is another important layer.

A Mineral Resource is not simply all the mineralised material that can be modelled.

Under JORC, every Mineral Resource must have reasonable prospects for eventual economic extraction (RPEEE), irrespective of whether it is classified Inferred, Indicated or Measured.

JORC describes a Mineral Resource as a realistic inventory of mineralisation which, under assumed and justifiable technical, economic and development conditions, might eventually become economically extractable. Material without those reasonable prospects cannot be included (JORC, 2012).

This means resource readiness also begins to touch areas that exploration teams may historically have regarded as something to worry about later.

Mining assumptions matter. Metallurgy matters. Geotechnical information matters. The likely geometry of extraction matters.

Metallurgical samples need to represent the deposit too

JORC requires the Competent Person to consider potential metallurgical methods when assessing RPEEE, even though the assumptions at Mineral Resource stage may still be preliminary (JORC, 2012).

That does not mean every early Mineral Resource requires a full feasibility-level metallurgical program.

But if metallurgical testwork is being relied upon to support assumptions about recovery, it needs to have some relationship to the material being reported.

A few convenient samples from one shallow, high-grade part of a deposit may not tell us much about the behaviour of a large, vertically and laterally variable mineralised system.

Leading international mineral-processing guidance recommends that metallurgical samples represent the major geological, mineralogical and metallurgical domains, have spatial coverage appropriate to deposit size and complexity, and include relevant grade ranges. It also recommends considering material around cut-off grade and potential dilution or gangue material, not simply the "best-looking" mineralisation (CIM, 2022).

In practical terms, the testwork should have a reasonable spread through the deposit: shallow and deeper areas, relevant mineralisation domains and, where appropriate, material around the mineralised boundaries that may ultimately enter the mining and processing stream.

Again, this is not about a prescribed number of samples.

It is about representativity.

And then comes the mining constraint

This is where projects can get a surprise.

JORC says that determining RPEEE requires consideration of possible mining methods, minimum mining dimensions and dilution assumptions. A Mineral Resource is specifically not simply an inventory of everything drilled above a selected cut-off grade (JORC, 2012).

In my view, the days of assuming that every block above a cut-off can simply be declared as a Resource are largely behind us.

That does not mean the JORC Code mandates a pit optimisation or a Mineable Shape Optimiser (MSO) for every Mineral Resource.

It does not.

But where open-pit or underground extraction is the reasonable mining assumption, some defensible demonstration of mineability becomes increasingly important.

A pit shell, mineable stope shapes or another appropriate constraint may show something that the unconstrained geological model did not.

The material you thought you were going to report may simply not fall inside the realistic extraction envelope.

Suddenly:

geological inventory ≠ Mineral Resource inventory

That can be quite a moment.

Industry discussion around RPEEE has highlighted exactly this issue: the reported Resource should reflect realistic mining and economic considerations rather than simply a grade-based inventory, and economic pit constraints or potentially mineable underground shapes are widely used ways of demonstrating this where appropriate (Glacken, 2019; CIM, 2019).

Mining engineers are very good at finding practical solutions. As I sometimes think of it, they can bend a piece of steel. But they cannot bend it forever. If geometry, minimum mining dimensions, dilution, geotechnical constraints or economics remove material from the realistic extraction envelope, no amount of optimism in the geological model puts that material back into a defensible Resource.

When you discover all of this at the end

This is where the real cost appears.

A company may have spent several months drilling, with years of historical data available. Thousands of metres have been completed. Assays received. Exploration models built. Budgets approved. Project-growth expectations established.

Then the Mineral Resource process starts.

Only now does the detailed review show that collars need resurveying, survey data need reconciliation, density coverage is inadequate, logging requires validation, metallurgical information is poorly distributed, geotechnical information is insufficient, or that the realistic mining constraint captures substantially less material than expected.

The mineralisation has not disappeared.

A lot of good exploration work may genuinely have been done.

But the project may still not be resource-ready.

And now it may need to move backwards before it can move forwards.

That can mean:

• resurveying;
• relogging;
• additional density work;
• metallurgical sampling and testwork;
• database remediation;
• new geotechnical work;
• infill or confirmatory drilling;
• reinterpretation;
• re-estimation; or
• reconsidering the Resource classification or even the amount of material that can legitimately be reported.

By then, the cost is no longer simply technical.

Schedules move. Consultants repeat work. Programs are redesigned. Budgets change. Management and board decisions may need to be revisited.

The cost of fixing the original problem can become much smaller than the cost of revisiting everything that was already built on top of it.

Beyond the model

The pathway ultimately looks something like this:

Data → confidence in the data → interpretation → model → estimate → classification + RPEEE → Mineral Resource → project decisions

And that is why resource readiness should not begin when someone decides it is time to produce an MRE.

It should begin while the project is still drilling.

The question during exploration should not only be:

Are we finding good mineralisation?

It should also be:

If this discovery becomes a Mineral Resource, are we collecting, validating and documenting today what we will need to defend tomorrow?

Finding mineralisation is one milestone.

Building a model is another.

Demonstrating that the material can defensibly become a Mineral Resource is another again.

And the earlier those three things are connected, the less likely years of hard work are to reach the end of the process only to discover that the project needs to go backwards before it can move forwards.

---

Note: The views expressed in this article are the author's own. References have been cited and applied to the context as accurately as possible.

Abzalov, M. Z. (2013). Measuring and modelling of dry bulk rock density for mineral resource estimation. Applied Earth Science, 122(1), 16–29. doi:10.1179/1743275813Y.0000000027.

Canadian Institute of Mining, Metallurgy and Petroleum. (2019). Estimation of Mineral Resources and Mineral Reserves Best Practice Guidelines. CIM Mineral Resource and Mineral Reserve Committee.

Canadian Institute of Mining, Metallurgy and Petroleum. (2022). Leading Practice Guidelines for Mineral Processing. CIM Sub-Committee on Leading Practice Guidelines for Mineral Processing.

Glacken, I. M. (2019). The highly vexed issue of reasonable prospects for eventual economic extraction (RPEEE) — Narrowing the range of practice. In Proceedings of the 11th International Mining Geology Conference. Australasian Institute of Mining and Metallurgy.

Joint Ore Reserves Committee. (2012). Australasian Code for Reporting of Exploration Results, Mineral Resources and Ore Reserves: The JORC Code, 2012 Edition. Australasian Institute of Mining and Metallurgy, Australian Institute of Geoscientists and Minerals Council of Australia.

Pressacco, R., Landry, P., & Evans, L. (2024). Some common flaws encountered in mineral resource estimation and how to avoid them. CIM Journal, 15(3), 172–200. doi:10.1080/19236026.2024.2322391.

The JORC Code 2012 Edition remains the operative Code at the time of writing. JORC has announced a further update on the provisional replacement for 20 August 2026.