BIOLOGIX NZ BLOG

Wrong Organic Matter, Wrong Calcium: How a Carbon Test Writes the Lime Order

Daniel Schuurman  ·  Biologix NZ  ·  24 September 2026

A soil test can put the right calcium in the wrong percentage. The kilogram of exchangeable calcium in the bag did not change. The denominator did.

Cation exchange capacity is that denominator. A large share of CEC sits on humus, not on clay. Get organic matter wrong and you get CEC wrong. Get CEC wrong and every base-saturation figure on the report moves — calcium first, because calcium occupies most of the seats. The lime recommendation is written from that percentage, not from the raw ppm. That is how a carbon method becomes a calcium programme, and how New Zealand paddocks have taken more calcitic lime than the colloid could hold.

This sits next to the companion article on excess calcium tightening soil and lowering oxygen. That piece is what happens after the truck has been. This piece is why the truck was called.

The only equation that matters

Calcium base saturation is not a lab measurement. It is a calculation:

% Ca  =  (exchangeable calcium in meq  ÷  CEC in meq)  ×  100

Exchangeable calcium comes off the colloid in ammonium acetate. CEC is either measured in the same extract (cations plus extractable acidity) or estimated from clay and organic matter. Either way, organic matter is in the story. Humus carries roughly 150–400 meq/100 g. Kaolinitic clay might carry 5–15. A 1% error in organic matter is not a rounding issue. It is extra parking spaces invented on paper, or parking spaces deleted on paper.

Figure. The calcium in the extract can be correct and the percentage still be wrong. That is a CEC problem, and CEC is an organic-matter problem as soon as someone estimates it.

A rule of thumb used in soil-fertility teaching makes the size of the error obvious. For every pH unit above 4.5, each percent of organic matter contributes about 1 meq of CEC per 100 g of soil. At pH 6, 4% OM is about 6 meq from humus alone. Report that same soil as 7% OM and you have just added 4.5 meq to the denominator. Calcium that was 57% of a 14 meq CEC becomes 43% of an 18.5 meq CEC. Nothing in the paddock moved. The report now says the soil is severely short of calcium.

Figure. Same 1,600 ppm exchangeable calcium. Two organic-matter figures. Two lime orders. The overstated OM is the one that fills the colloid.

Lime software does the next step automatically. New Zealand labs have long calculated a lime rate from the calcium needed to reach a target total base saturation — commonly around 75% on a buffered CEC. Albrecht-style programmes target about 68% calcium specifically. Both are percentages of CEC. Inflate CEC, and the tonnes required to get there inflate with it.

The error also runs the other way. Understate organic matter, shrink CEC, and calcium looks finished when the soil still needed it. That is how one block is over-limed and the next is left acid, from the same habit of trusting a single OM bar.

NIR is a prediction. Dumas is a measurement.

Labs do not all mean the same thing when they print organic matter percent.

Dumas combustion is the reference method for total carbon. The dried, sieved soil is burned. Carbon leaves as gas and is measured as CO2. Organic matter is then calculated with the van Bemmelen factor: OM % = total C % x 1.724, on the assumption that humus is 58% carbon. That factor is itself an approximation — New Zealand soils are treated as low-carbonate, so total C is used as if it were organic C — but the carbon number is a physical measurement of the sample in the boat.

Near-infrared spectroscopy (NIR / NIRS) is not chromatography, and it is not combustion. It is a secondary method. The instrument records how the sample reflects light in the near-infrared. A calibration model, built on soils that were also run by a reference method, predicts total carbon or organic matter. The prediction is only as good as the library and the lookalikes in it.

That is the reliability gap. Reviews of vis-NIR for soil organic carbon typically land around a validation R2 near 0.67, with prediction error that is often one-and-a-half to two times the error of dry combustion on the same sample. Combustion remains more accurate on the individual sample. NIR wins on speed and cost. Those are different jobs.

New Zealand commercial practice already admits the gap. One major laboratory runs agricultural total carbon by NIRS as the default, then code-swaps samples it cannot predict cleanly across to Dumas. Their own NIRS technical note puts the accepted-prediction rate around 85%, with the rest sent to wet chemistry because uncertainty of measurement is higher. Volcanic ash, peats, recent lime, high carbonate, unusual colour and soils outside the calibration neighbourhood are exactly where a spectrum stops being a number you should build a lime rate on.

Two further traps sit on top of the instrument.

First, total carbon is not cation-exchange humus. Dumas and a well-calibrated NIR both see charcoal, fresh residue, roots and, if carbonates were not removed, lime carbon. Only the humified fraction carries the bulk of the organic CEC. A high OM that is mostly unhumified carbon inflates estimated CEC without adding exchange sites. Percent calcium falls. Lime is recommended for parking spaces that do not exist.

Second, the 1.724 factor assumes a carbon content humus does not always have. Using total C as organic C on a recently limed or shell-influenced soil overstates OM. An NIR model trained on mixed methods (Walkley-Black, loss-on-ignition, combustion) will inherit those biases and print them as one tidy percentage.

Ask the lab which method produced the OM or total C on the page. If the answer is NIR only, and you are about to calculate base saturation or a lime rate from CEC that used that figure, pay for Dumas on that sample. Trends in carbon over years can live on NIR. A lime order should not.

Why this became extra calcitic lime

The carbon error does not arrive alone. It arrives in a programme that already prefers calcium to magnesium.

For twenty years the nutrition-farming reflex has been: raise pH, widen Ca:Mg, flocculate the clay. Calcitic lime does all three in one pass. Magnesium is treated as the ion that tightens soil — which it does, in excess — so dolomite is left in the shed even when the leaf is short of Mg and the cow is in the tetany window.

That reflex has a cost the Calcium module already teaches.

•    Magnesium has about 1.4 times the liming effect of calcium on an equivalent basis. Using only calcite to chase pH ignores the cheaper pH work magnesium would have done, and ignores the plant and the rumen that still need it.

•    Large calcium inputs displace magnesium on the colloid. The course rule of thumb is about 1% magnesium base saturation lost for every 1% calcium you lift.

•    The target Ca:Mg ratio is real for structure. It is not a reason to drive calcium to 80% while magnesium falls through 10%. That is how you leave the working window: free carbonate, high pH, slaked crumbs, less oxygen.

Stack the two mistakes and the pattern we keep seeing is no longer mysterious.

•    NIR or a mixed OM method overstates carbon, or treats raw carbon as humus.

•    Estimated CEC rises. Calcium base saturation prints low.

•    The report, or the consultant software, calculates lime to a 68-75% target.

•    The product chosen is calcitic lime, because the Ca:Mg ratio also looks too narrow on that same inflated CEC.

•    Magnesium is displaced. pH jumps. The next test shows even more calcium and even less magnesium — or a leaf short of both.

•    Another lime pass is booked, because pH or %Ca still is not there, and nobody has asked whether the denominator was ever true.

Figure. The test error writes a low %Ca. The product choice fills the seats with calcium and knocks magnesium off. Together they produce the excess-calcium soil the last article described.

Measured CEC is not immune. Buffered CEC at pH 7 includes latent acidity that organic matter helps generate. Summation CEC on a soil with free lime extracts carbonate calcium that was never on an exchange site, inflates both Ca and CEC, and scrambles the percentages. High pH and free CaCO3 — the end-state of the over-liming habit — are exactly when the cation maths is least trustworthy. That is another reason a high-pH, high-Ca report should not be answered with more carbonate.

What to do with the next bag of soil

The Biologix soil-testing protocol does not change: 150 mm cores, exchangeable cations by ammonium acetate, Mehlich 3 and TAE phosphorus, and a report read against Biologix ideals rather than a generic pasture bar. What this article adds is a rule for the carbon line.

•    Ask how OM and total C were measured. Dumas combustion for the sample you will lime from. NIR is acceptable for monitoring carbon year to year once you have a Dumas baseline on that transect.

•    Read exchangeable calcium in ppm and meq before you read %Ca. If the ppm is already in range for that CEC class and the percentage looks desperate, suspect the denominator.

•    Do not estimate CEC from an NIR organic-matter figure and then calculate lime from that CEC. Use a measured CEC from the same cation extract, and treat high-pH, high-carbonate CEC with suspicion.

•    Match the lime type to the cations, not only to pH. Calcitic lime when calcium is short and magnesium is adequate. A lime-dolomite mix when both are short. Gypsum when pH is already high and magnesium or sodium is the packer. Foliar magnesium when the soil Mg is already high and the leaf is empty.

•    Stop before the printout says you have arrived. There is a delay between lime and the base-saturation lift. The Earthworkers cap on a single calcium lift exists because overshoot is expensive and tight.

•    Pair soil with leaf or sap. A high soil calcium and a low leaf calcium is uptake and antagonism, not a shortage of lime. A low soil %Ca and a comfortable leaf Ca on a high-OM report is a reason to check the method, not to book another truck.

On sands with CEC below about 5, ignore the percentages altogether and work in ppm. Organic-matter error on a sand still matters for carbon and water, but it should not be allowed to invent a base-saturation drama.

What this is not

It is not a claim that NIR is useless. It is a fast survey tool. It is not a reference carbon number, and it is not a licence to skip Dumas when the next decision is tonnes of lime.

It is not a claim that every New Zealand laboratory is inventing CEC from NIR. Many report a measured, buffered CEC from the ammonium-acetate extract. The failure starts when that CEC, or a consultant's estimated CEC, is then treated as gospel while the OM method sitting next to it would not survive a re-run by combustion.

It is not a reason to stop liming acid, aluminium-active soils. Those blocks still need calcium and a pH that lets traces and biology work. They need the right rate and the right calcium-to-magnesium mix, calculated from numbers that were measured.

The point

Organic matter is not a side panel on a fertility test. It is part of the arithmetic that decides how much calcium the report says you are missing. NIR predicts carbon from a spectrum. Dumas measures it. Use the prediction for a trend. Use the measurement when you are about to change the colloid.

Then stop using calcitic lime as the default answer to both pH and Ca:Mg. Magnesium is not an obstacle to a good ratio. It is half of the ratio. Drive only calcium and you will get the soil the last article described: high base-saturation calcium, displaced magnesium, free carbonate, a tighter profile and less oxygen.

Check the method on the carbon line. Read ppm and percentage together. Choose the lime that the cations actually need. That is how a soil test changes the next decision instead of repeating the last one.

Written for the Biologix NZ blog from Earthworkers 201 soil-testing and calcium teaching, New Zealand laboratory method notes on NIRS, Dumas total carbon and buffered CEC, and the companion article on excess calcium and soil structure.

Further reading

•    Earthworkers 201 Soil and Leaf Testing module; Calcium module (lime / pH non-response)

•    Hill Laboratories technical notes: Cation Exchange Capacity and Base Saturation; Laboratory Tests for Soil Carbon; Analysis of Soils Using Near Infra-Red Spectroscopy (NIRS)

•    Magdoff and van Es on organic matter contribution to CEC

•    Companion Biologix article: Excess Calcium Tightens Soil and Lowers Oxygen

Next step: submit samples through the Biologix soil and leaf reporting service (biologix.co.nz). Ask that total carbon be run by Dumas when the report will drive lime, and that exchangeable cations and CEC come from the same ammonium-acetate extract. Pair with a leaf or sap test before another calcitic pass.