Anything resembling thatch or mat is explicitly excluded from the soil organic matter measurement made on routine soil nutrient analyses.

That portion of soil organic matter is excluded because it is not measured. Those big chunks of organic matter are removed from the sample prior to testing by passing the sample through a 2 mm sieve. And there is a good reason for that.

This small material that does get tested is humus, and it meets the definition of soil organic matter in the Soil Science Society of America’s glossary.

soil organic matter: The organic fraction of the soil exclusive of undecayed plant and animal residues. See also humus.

humus: The well decomposed, more or less stable part of the organic matter in mineral soils.

My previous post concerned the spatial variability of total organic matter in the top 2 cm of putting green soils. That’s from OM246 measurements, which measure the total organic matter in the sample by—no surprise—measuring the entire sample. There’s no removal of undecayed plant and animal residues.

The coefficient of variation (cv) that I would expect in the total organic matter right at the soil surface in the top 2 cm, going from point to point across the green, is 14%. That means that on average, I’d expect the next point I sample to be 14% higher, or lower, in total organic matter, compared to the average total organic matter measured across the entire green.

multiple samples collected from the same green

What about the humus? What’s the spatial variability of that? I’ve been doing some research about this too, and I checked the cv of samples taken from the same green. These data are from samples taken to a 10 cm depth (4 inches), with the undecayed material removed. This is the organic matter you see on regular soil tests—this is the humus. This is the number to look at for estimates of nitrogen mineralization and for calculating the contribution of soil organic matter to the soil CEC. For evaluating sand topdressing requirements, or assessing sand topdressing effect, I’d use OM246.

This is a small dataset, with 75 samples so far from six greens. The median cv for these is 13%. It’s almost the same as the variability in the total organic matter at the surface.

For more about this, see: