How Moisture in Mineral Samples Corrupts Grade and Valuation

A mineral sample can weigh correctly, grade correctly, and still be wrong in a way that costs millions. Moisture is the reason. It rarely receives the scrutiny the problem actually warrants.

Moisture content is one of the most routinely underestimated variables in mineral sample management. It alters measured weight, dilutes or inflates apparent grade, changes how a sample behaves during crushing and splitting, and shifts the reliability of every model built on those numbers. The consequences are not theoretical: a 0.1% moisture error across a single year of Australian iron ore exports equals roughly US$35 million in misstated value.

What follows gives you a clear understanding of what moisture actually does to a sample, why the distinction between free and chemically bound water matters for measurement, and what the financial stakes look like when those measurements go wrong.

Why moisture changes what a sample appears to be

Start with the intuitive part. A wet sample weighs more than the same sample dried out, and that extra weight is nothing but water.

The trouble begins the moment grade enters the calculation. Grade is usually expressed as a percentage of total mass, so when the total mass is inflated by water, the calculated grade is diluted. Wet ore looks poorer than it is, or richer than it is, depending entirely on which moisture assumption you feed into the sum.

That directional problem is worth sitting with. If a system assumes lower moisture than is actually present, apparent grades inflate. If it assumes higher moisture, grades deflate. The error is not random noise; it is a systematic bias whose sign you control through your assumption.

Consider a documented density-based estimation case. A 25-tonne iron ore bucket carrying 2% moisture, with a true iron grade of 50.71% Fe, was reported as 54.10% Fe when the wet mass was treated as dry mass.

A 3.39 percentage-point overestimation, driven by nothing more than uncorrected moisture on a single truckload.

That figure tells you something uncomfortable: a modest 2% of water is enough to make materially different ore look like high-grade ore. Every purchasing and processing decision resting on that number is compromised before it is even made.

Iron ore grade manipulation, whether deliberate or the product of systematic measurement failure, produces the same commercial and regulatory consequences: royalty shortfalls, disputed shipment values, and eroded trust in the numbers underpinning contracts.

The distortion does not stop at grade. Moisture reaches into three distinct places:

• Grade dilution or inflation. Water in the total mass shifts the calculated percentage of valuable mineral up or down depending on the moisture assumption applied.
• Cross-sample comparison failure. Samples collected under different weather, depths, or stockpile positions carry different moisture loads, so comparing them directly is unreliable unless each is corrected to a consistent dry basis.
• Process model misalignment. Metallurgical models tuned to a fixed moisture value mis-predict behaviour when the real value drifts.

That last point has hard numbers behind it. In copper ore dry grinding, classifier performance degrades noticeably once moisture climbs above 1.5-2%. Measurement instruments feel it too: 10% moisture in iron ores produces a 4% relative error in iron grade measured by X-ray fluorescence (XRF), a technique that reads elemental composition, unless the result is explicitly corrected using scattered radiation intensities.

So moisture is not a nuisance you tidy up at the end. It reframes every downstream number, which is why it deserves the same rigour as any assay.

Free water, bound water, and why the difference matters for measurement

Here is a deceptively simple question: what kind of water is actually in your sample? The answer determines which drying protocol is correct, what temperature to use, and whether drying itself damages the mineral you are trying to measure.

There are three categories, and they behave very differently.

Free (evaporable) moisture is water sitting on particle surfaces and in pores, in excess of equilibrium. It matters for handling, flowability, and dust suppression, but it is not chemically attached to the mineral structure.

Bound (hygroscopic) moisture is water held tightly in thin films on surfaces and inside fine pores. Standard oven drying at around 105 degrees Celsius removes it without disturbing the mineral, which is exactly why ISO analytical standards target it when correcting chemical grades to a dry basis.

Crystalline (combined) water is chemically integrated into the mineral lattice itself, in clays and hydrated oxides. Removing it demands far higher temperatures, often above 200-400 degrees Celsius, and its release comes with structural change. ISO 7335 defines combined water as the water liberated between 105 plus or minus 2 degrees Celsius and 950 plus or minus 20 degrees Celsius.

This is where “just dry it” stops being a complete answer. ISO 2596:2006, for instance, targets hygroscopic moisture, typically 0.05-4.5%, by heating at 105 plus or minus 2 degrees Celsius for up to 24 hours. Push the temperature higher to chase every last molecule and you start attacking the crystalline water, corrupting the very grade you meant to correct.

What happens when the wrong drying protocol is applied

Apply a blanket 105-degree-Celsius protocol to the wrong material and you do more than mismeasure. You change the sample.

Drive gypsum above roughly 105 degrees Celsius and it dehydrates to bassanite. High-sulfide ores risk phase transformation and oxidation. This is precisely why CIM guidelines insist on temperatures low enough to avoid driving off water of crystallisation in lateritic deposits. Once that structural change happens, any subsequent analysis of the sample is meaningless.

Then there is the quieter failure of re-adsorption. If samples cool in humid air without airtight containers, they take moisture back on unevenly before their mass is ever recorded, which invalidates the measurement. Both CIM and ISO guidance handle this by instructing analysts to accept the mass recorded just before any increase, because a rising mass on repeated drying signals re-adsorption or decomposition rather than genuine water content.

What the financial stakes actually look like

Start small. A 0.1% moisture error on 1 million tonnes of iron ore priced at US$50 per tonne shifts the cargo’s stated value by US$50,000. On a single shipment, that is already real money.

Now scale it. Australia exports roughly 700 million tonnes of iron ore per year. Hold that same 0.1% error rate across the national flow and the arithmetic becomes striking.

A 0.1% moisture error across Australia’s annual iron ore exports equates to approximately US$35 million a year in misstated value.

That figure reframes moisture entirely. It is not a laboratory compliance exercise; it is a commercial accuracy problem operating at national commodity-flow scale. Producers who assume moisture rather than measure it are quietly accepting that level of valuation uncertainty as the cost of doing business.

The exposure runs through three distinct pathways:

1. Cargo-level valuation error. Small percentage errors on large tonnages translate directly into six- and seven-figure swings in stated value.
2. Metallurgical metal content accounting error. A 1% relative error in moisture content produces a corresponding 1% relative error in calculated metal content, so moisture inaccuracy propagates straight into your metal balance.
3. Commodity quality threshold misstatement. Coal nominally sitting at 30% moisture typically ranges between 27% and 33% (plus or minus 3 percentage points) 95% of the time; assume a single value and you inject several-percent errors into dry tonnes and energy content.

The consequences reach beyond spreadsheets. In one platinum operation, free water seeped from concentrate bags during transport, changing apparent mass in transit. Because moisture factors were assumed rather than representatively measured, tonnage and metal balances stopped reconciling, and plant performance assessments were misled by numbers that were never wrong in the lab, only wrong in what they assumed.

That is the pattern worth remembering. Moisture measurement is not a technical footnote to chemical assay. It carries equivalent financial weight and, too often, a fraction of the rigour.

Measurement standards and the protocols that govern them

The protocols governing moisture look, at first glance, like a compliance checklist. They are not. Every temperature threshold and container specification is a deliberate response to a documented way the measurement fails.

Take ISO 3087:2020 for iron ores. It requires test portions spread in layers no deeper than 31.5 mm, dried at 105 degrees Celsius for at least 4 hours, until successive weighings change by no more than 0.05% of the initial mass. The layer depth stops water being trapped under a thick pile; the constant-mass criterion stops you calling the job done before the sample is genuinely dry.

ISO 3087:2020 specifies the method for determining moisture content across a lot of iron ore, covering natural and processed material, and its constant-mass criterion is the direct technical basis for the layer depth and weighing interval requirements applied in commercial iron ore transactions.

Each standard operationalises the free, bound, and crystalline distinction in its own way.

Handling standards address the moment before the lab even receives the sample. The IMSBC Code from the International Maritime Organization (IMO) requires moisture samples to be sealed immediately in airtight, non-absorbent containers with minimal free air space, precisely because evaporation and re-adsorption corrupt the result in transit.

A practical protocol resolves the underlying tension. In coal and ore operations, a dual-sample approach uses one sealed sample for total moisture and one unsealed sample for analysis, then mathematically corrects the analytical results back to field moisture. You preserve the real moisture value while still running the analytical preparation the lab needs.

Reporting codes and the disclosure obligations they impose

The regulatory architecture goes further than measurement. It treats moisture basis as a material assumption you must disclose.

Under the JORC Code 2012, a Competent Person must state whether tonnages are estimated on a dry basis or with natural moisture, and explain how moisture content was determined. The CRIRSCO International Reporting Template (2019 and 2023-2024 updates) similarly requires disclosure of the moisture basis used for relative density and encourages cross-validation of wet and dry sample weights against hole volume and moisture factors. CIM best practice adds the spatial dimension, documenting how moisture is distributed through a deposit, particularly in lateritic profiles where it varies widely.

Supergene weathering produces the lateritic profiles that CIM guidelines specifically flag as high-moisture-variability zones, where clay minerals and hydrated oxides hold crystalline water that standard 105-degree protocols cannot safely remove without altering the sample’s mineralogy.

What this tells you is direct: if a Competent Person cannot state how moisture was determined, the tonnage estimate resting on that assumption is not considered fit for public reporting.

The cost of under-measuring surfaces plainly at Brazil’s Pico complex. Moisture estimated from normative mineralogy came in roughly 2% below the measured product moisture. Because the critical limit for the dry beneficiation plant is 12.5%, planners had to restrict operations to 10.5% moisture to build a safety buffer. That 2% estimation gap translated directly into reduced throughput.

Where the measurement breaks down in practice

Knowing what the standards require is one thing. Recognising where they quietly fail in your own operation is another, and moisture error tends to enter at specific, identifiable handoff points rather than as an abstract possibility.

The failure modes show up repeatedly across commodities:

• Bag dewatering before measurement. In a documented case, pulp samples stored in non-leak-proof big bags were allowed to dewater before moisture was measured, introducing an enormous bias into moisture estimation and the metal reconciliation that followed.
• Assumed rather than measured moisture. In the platinum operation noted earlier, moisture factors were assumed instead of representatively sampled, so tonnage and metal balances appeared inconsistent and misled plant performance decisions.
• Unmeasured moisture in geotechnical modelling. Numerical modelling of an underground limestone mine showed that unmeasured or underestimated moisture in floor materials reduced pillar strength below critical stress limits, a stability consequence well beyond grade and accounting.

The dewatering case is the instructive one. It shows that moisture error needs no calibration fault and no equipment failure. It can enter the system through a single storage decision made between the field and the laboratory, which means discipline at handoff points matters as much as protocol at the bench.

Moisture is one variable in a broader chain where laboratory integrity determines whether the numbers reaching decision-makers reflect reality; calibration drift, sample contamination, and chain-of-custody failures can each compound the moisture errors introduced before a sample reaches the bench.

Six safeguards form a practical operational framework:

1. Standardised protocols. Hold to defined temperatures, typically 105-110 degrees Celsius, and constant-mass criteria.
2. Low-temperature drying for sensitive materials. For clay-rich, high-sulfide, or hydrated minerals, dry at 45 degrees Celsius or below to avoid structural change, even if it takes several days.
3. Airtight preservation. Seal moisture samples immediately in airtight, non-absorbent containers with minimal free air space.
4. Dual-sample approach. Keep one sealed sample for total moisture and one unsealed for analysis, correcting analytical results back to field moisture.
5. Frequent, representative measurement. Measure moisture regularly in ore feeds and stocks rather than leaning on historical values.
6. Explicit reconciliation tracking. Treat moisture basis as a core reporting assumption, tracking dry tonnes and moisture factors across drilling, mining, processing, and shipping.

A reader who can name these failure modes is equipped to audit their own sample chain rather than assume moisture is being handled correctly somewhere downstream.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.

Making moisture count before the decisions that depend on it are made

Pull the thread together and the shape is clear. Moisture is not a correction bolted on at the end of analysis. It is a variable that conditions every weight, grade, and processing figure from the moment of sampling onward, and it does so whether or not anyone measures it.

The obstacle is rarely technical. The standards are well established, the temperatures are specified, the protocols are documented. What lets moisture slip is procedural and cultural: it tends to receive less rigour than chemical assay despite carrying equivalent financial and interpretive weight, as the US$35 million annual figure and the Pico throughput restriction both demonstrate.

So the decision in front of you is not whether to measure moisture. It is when, how, and under which standard, and whether the reporting basis, dry or natural moisture, is stated explicitly at every stage of the workflow where a number is communicated. Get that discipline right and the numbers you act on become numbers you can trust.