Engineering & Technology · Mining, Resource Recovery & Geoengineering Systems

Resource characterisation, uncertainty and geological evidence

A laboratory can analyse a small sample carefully while the surrounding deposit remains poorly known. Resource characterisation asks both whether the sample result is sound and what volume of ground it can reasonably represent. Precision in one assay cannot replace evidence about geological continuity.

Use Geotechnical & Ground Engineering, Process Balances & Design and Reliability, Safety & Maintenance for ground response, inventories and complete-system dependencies. Physics continuum materials and Chemistry separation science provide additional preparation. This chapter supplies process principles and evidence questions, not extraction, support, ventilation, chemical-treatment or reservoir-operating procedures.

Back to the chapter

After reading, you can

Connect geological observations to an explicit model of volume, density, grade and uncertainty, keeping contained material separate from recoverable product.

Keep observations and interpretations distinguishable

Maps, borehole records, exposed rock, geophysical responses and laboratory analyses provide different evidence. Record locations, depths, sample intervals, recovery and preparation history. An element concentration does not by itself identify the mineral that contains it, and the mineral host can strongly affect whether a useful separation is possible.

A geological model connects observations through interpreted boundaries, structures and mineral relationships. A block between holes is usually an estimate, rather than a fully observed object. Repeat analyses, reference materials and sample custody address measurement quality; additional geological coverage addresses a different uncertainty. Several nearby samples may also share information rather than constitute independent evidence.

Grade needs a stated material and mass basis

For this chapter, metal grade means mass of a named element divided by dry bulk mass. The element can be chemically bound in a mineral; “contained metal” does not mean that a piece of pure metal is present. Moisture, density and the selected boundary affect the conversion from volume to dry tonnes.

An estimated mineral resource and material judged economically extractable under a defined plan are different claims. Economics also needs recovery, access, infrastructure, environmental and social conditions and other modifying evidence. Formal resource and reserve categories depend on the applicable reporting framework; these lessons do not assign a category or supply an investment disclosure.

Weight grade by the quantity actually being averaged

Imagine two homogeneous model zones, with all inputs supplied rather than measured. Zone A has volume 4000 m³, dry bulk density 2.5 t/m³ and target-element grade 0.8%. Zone B has 6000 m³, density 3.0 t/m³ and grade 1.2%. Their dry masses are 10000 and 18000 t, containing 80 and 216 t of the target element.

Combined dry mass is 28000 t and contained target mass is 296 t, so combined grade is 296/28000≈1.0571%. A simple average of the two percentages gives 1.0%, while volume weighting gives 1.04%. Neither uses the required mass weights when zone masses differ. The figure shows the declared geometry behind this arithmetic.

A slab has supplied cross-sectional area 1000 m² and lengths A 4 m, B 6 m, drawn at 50 drawing units/m. A volume 4000 m³, density 2.5 t/m³, grade 0.8% gives 10000 t dry material and 80 t target. B volume 6000 m³, density 3.0 t/m³, grade 1.2% gives 18000 t and 216 t. Combined grade 296/28000=1.057142857%. Rectangles show zone lengths, not individual mineral grains or an inferred geological boundary.

Change one uncertain input without inventing a confidence interval

Suppose only Zone B’s supplied grade is varied from 1.2% to 1.0%, leaving volumes, densities and Zone A unchanged. Its contained target mass becomes 180 t. The combined result is then 260 t and grade 0.9286%. That is a sensitivity case, not an established lower confidence bound or evidence that every intermediate value is equally likely.

The original 296 t is contained inventory in the defined model. Mining losses, dilution by other material and processing losses have not been applied. Report the model version, evidence and assumptions with the result, and identify which new observation would most effectively challenge the interpretation rather than adding unsupported decimal places.

Try the reasoning

Why is the simple mean of 0.8% and 1.2% inappropriate here?
Show the answer and reasoning
The zones have unequal dry masses. Combined metal grade must weight each grade by its corresponding mass.
Does a carefully repeated assay establish the grade between boreholes?
Show the answer and reasoning
No. It helps assess analysis of that sample; spatial representativeness and geological interpretation need separate evidence.
Is 296 t the amount of saleable metal?
Show the answer and reasoning
No. It is contained target-element mass in an invented geological model before extraction, dilution, processing and product constraints.

What to carry forward

Preserve the chain from observations to interpretation, and state the mass basis, uncertainty and recovery boundary of every quantity.

References

Teaching resources and research supporting this lesson’s explanations and examples.