Share This:

Soil sampling

Last updated: 06 Dec 2016

Soil sampling is one of those jobs that’s simple in principle but tricky to get right in practice. As phosphate and potash recommendations (and so fertiliser costs) are based on soil indices, correct sampling is absolutely essential and it pays to do it properly. Sampling procedure is the prime suspect whenever an index changes suddenly or you get an unexpected result in a soil analysis report.

Everyone knows about the ‘W’ pattern for taking 25 or more soil cores across the field (a grid pattern can be just as good but involves a bit more walking). It’s used to try to ensure the cores, when bulked into the sample for analysis, give a result that is representative for the main cropped area of the field. Unusual small areas need to be avoided.  These tend to be fairly obvious like headlands and areas around gates or feeders but they include sites of previous manure heaps and old hedgerows which might not be visible or obvious.

If fertiliser is to be spread at the same rate across the field, one representative sample is the best you can do. However, if the field is large and there are differences in soil type across it, that are large enough to allow different fertiliser spreading rates, it will be better to sample each of these areas separately. This puts the analysis cost up but as typically it’s only around 25p/ha/year, it’s worth it.

The sample sent to the laboratory will be dried, ground and mixed thoroughly before analysis. However, if the bulked sample taken in the field is too large to send to the laboratory, it will need to be sub-sampled. The bulked sample must be really thoroughly mixed to make sure the sub-sample sent to the laboratory is representative. The result you get is only as good as the sample that reaches the laboratory.

Soil phosphorus and potassium levels are reported in milligrams per litre or parts per million. In more familiar units, this is equivalent to one kilogram in 1000 tonnes. So just a small amount of contamination from fertiliser dust or a granule in the sample will affect results. Hands and tools should be clean before sampling and, when taking a core, contaminants like bird droppings or traces of granules must be avoided – not always as easy as it sounds on a miserable cold day when enthusiasm starts to run out on about the twentieth core taken.

Then there is the timing of sampling. Ideally, this should be at least six months after the last phosphate and potash application and, if possible, two years since liming. This is to allow fertiliser nutrients and lime time to mix with the soil and settle down to give a stable result. The best time is before base fertiliser is applied for the next crop so autumn for winter crops and autumn or late winter/early spring for spring sown crops.

Depth of sampling is important too. The standard for arable soil is 15 cm. This is rather less than plough depth but, allowing for the soil mixing in the plough layer, should give a result representative for this layer. Problems occur when lime or fertiliser is applied to the surface and is then ploughed down before soil sampling because the corer will miss the application and give a false result. This is the usual issue when liming seems not to raise soil pH. Problems also can occur where min-till with occasional ploughing is practised. Phosphate in particular is immobile in the soil and mixing depends largely on cultivation. Under min-till, phosphate will tend to remain near the soil surface but it will become mixed with the soil after ploughing. So, a sample taken to 15 cm will give different results depending when it is taken in relation to the ploughing (higher during min-till, lower after ploughing). The Potash Development Association has proposed sampling to full plough depth in this cultivation system to avoid this problem (

Soil sampling is not quite as easy as it sounds if it is to be done properly and to give the correct result. A bit of time in preparation and care in doing it are well worthwhile.

There are good detailed instructions for taking soil samples at the Tried & Tested web site (, in the Fertiliser Manual (RB209) ( and in the SRUC leaflet ‘How to take a soil sample’ (