Everything depends on the laboratory results being correct and representative for the field or area sampled. Two things are needed for this:
- the sample taken and sent ot the laboratory must be representative and,
- the sample must be analysed correctly and accurately by the laboratory
When things go wrong, the laboratory tends to get a kicking but is this fair? When indices change suddenly or unexpectedly, the temptation is to say the analysis must be wrong but in most cases, this turns out to be mistaken. In fact, steps were taken some years ago to ensure consistency and accuracy of laboratory results. The main UK laboratories that offer soil analysis formed the Professional Agricultural Analysis Group (PAAG) in 2009 and take part
in the independent, internationally recognised and longstanding proficiency testing scheme (ring test) run by Wageningen University in the Netherlands. A leaflet just issued by PAAG to explain this is now at the Tried & Tested website. The laboratories involved are shown in the leaflet. The scheme allows every participating laboratory to compare its results with those of other laboratories for sets of standard soil samples provided quarterly by Wageningen. This continuous monitoring would reveal any significant differences in results among laboratories but so far consistency has been good.
If the laboratory analysis is good, where can problems lie? Well, there’s much more scope for errors in soil sampling than there is in the laboratory. After all, the soil in a field of several hectares must be represented by the five grammes used for analysis. Soil sampling is a more likely cause of any odd results and is the first thing to check. Often any problem can be related to cultivations, especially soil inversion that buries surface applied lime, manure or fertiliser to beneath the standard 15 cm sampling depth. Changes from ploughing to minimum tillage also can affect results. It’s really important that good care is taken in sampling and field history as well as visible features must be taken into account. Everyone knows headlands and gateways should be avoided but sites of long-disappeared manure heaps can still be high in nutrients and a core taken from them will skew results.
Soil analysis is incredibly good value. For a typical cost of around 25p/ha/year for laboratory analysis, phosphate and potash use can be fine-tuned to meet the needs of the soil and crop and to support the nitrogen applied. But it does depend on a representative sample so when a result looks odd, first have a good look at how the sample was taken.