Soil is sampled on a grid and sent to a laboratory, and the numbers come back as a map. One field has a circle about eight metres across where the pH, the nitrogen and the phosphate are off every scale the lab uses, and the circle is in a different place every year.
I am a soil sampler for an agronomy firm. I have cored about forty thousand samples.
How sampling works.
You drive the field on a grid, twenty five metres or fifty depending on the job, and at each point you take a core to fifteen centimetres, bag it, and log the position from the GPS on the bike. The bags go to a laboratory that measures pH, phosphate, potash, magnesium and, on some jobs, organic matter and nitrogen.
The results come back as numbers against positions, and the software draws a map, and the map is the point of the job: the farmer applies lime and fertiliser to the map rather than to the whole field.
A field's chemistry changes slowly. Year on year the map looks like last year's map with small drifts, because soil is soil.
One field on one farm has a patch about eight metres across where the numbers are not soil.
pH nine point four. Phosphate index seven, which is off the top of the scale that farming uses. Nitrogen at a level the laboratory queried as a possible contamination.
Around the patch, and everywhere else in the field, it is ordinary loam at pH six and a half.
It is in a different place every year.
I have sampled that field annually since 2018 because the farmer is thorough, and the patch has moved about two metres a year, in a straight line, west to east, across the field.
Seven years, fourteen metres. It is now about a third of the way across.
What has been checked.
Contamination. This was the laboratory's answer and it was a fair one, so I re-sampled the patch with a clean corer on a clean day, double bagged, and sent it to a second laboratory under a different reference.
Same numbers.
The ground. pH nine is what you get from a lime spill or from concrete or from ash. There is no concrete, no spill, no ash. There is nothing in the field's history, which the farmer has back to his grandfather, and nothing on the old maps.
I have dug it. A pit, a metre down, at the centre of the patch, in 2021. Loam over clay, the same profile as a pit ten metres away, roots and worms and nothing else.
The chemistry in the pit is the patch's chemistry at every depth to the clay, and then it is ordinary.
The crop.
This is what the farmer cares about and it is the part that makes no sense in a different way.
Soil at pH nine with that much phosphate should grow either nothing or something very strange. It grows the crop. The same wheat as the rest of the field, the same height, the same yield on the combine's monitor, and I have walked it at every stage and you could not find the patch from the crop.
The soil says one thing and the plants say another and they are growing in it.
Where it is going.
Two metres a year, west to east. I have a plot. At that rate it reaches the eastern hedge in about twelve years, and beyond the hedge is the farmhouse garden.
I have told the farmer that, in those words, because he is entitled to know. He laughed and then he did not.
What is done about it.
The patch is excluded from the fertiliser plan, which means it gets the rate the rest of the field gets, because applying the rate the numbers ask for would be applying nothing, and applying nothing to a patch of a field that grows a normal crop is a decision nobody wants to make on paper.
The laboratory has both sets of results on file with a note that says anomalous.
What I do.
I sample it every year on the grid and I add a dense grid across the patch and twenty metres either side, so that I know where it is.
The one thing I have changed is that I sample that field first, in the morning, and not last.
The last field of the day on that farm used to be that one, because it is nearest the yard and it is the one you finish on and go home.
In 2022 I was in it at about half past five in September with the light going, and I took the core at the centre of the patch, which I always do last, and the corer went in like a corer goes into loam, and came out warm.
Not the soil. The steel. The tube of the corer was warm enough to notice through a glove, the way a tool is when it has been in the sun, and the sun was gone, and the soil in it was the temperature of soil, and the three cores I took either side of it were not warm at all.