Think Agronomy

By David Jones, CropNuts Broad Acre Agronomist

This week in THINK AGRONOMY…

  • Why soil sampling improves profitability
  • How to sample properly
  • Grid vs zone sampling

It’s The Planting Season

As we approach planting in many areas of Kenya, we are also approaching one of the largest expenditures of the year for most growers.

Our costings show that last year our Agronomy Service clients spent between Kes 4,300 and 16,200 per hectare on fertiliser for wheat. This is a vast range, but it also reflects the variation in soils, rainfall and yield potential across Kenya.

Land Preparation

It’s planting season in most parts of Kenya

Applied in the correct quantities, the potential of the crop is maximised and the actual cost per ton of wheat produced can be minimised as the table below shows; three different farms and different yields, but by soil sampling and creating individual fertiliser recommendations, profitability can still be maintained.

Fertiliser cost per ton

Getting the right amount of fertiliser on to maximise the return on investment is crucial, and depends on two key factors, yield potential and what is already in the soil. Yield potential and crop requirement is a matter of conjecture and amounts to a guess at the rainfall that is likely to follow.

Knowing what is in the soil on the other hand does not need to be guess work, but should be the result of a soil test. A reminder of the basics:

  • Soil sample every field at least once every year, or when you experience any unexplained changes in crop growth.
  • Sample to 20 cm, taking an even core and mixing the soil thoroughly before sending it to the lab.
  • In no till systems I periodically sample in layers of 5cm to look at changes in acidity. This has been a very useful of guide of liming effectiveness

Only as good as the sample

The sample result is only as good as the quality of the sample taken and delivered to the laboratory. Sample fields in a ‘W’ pattern across fields in order to take a representative sample across the whole field. This should be made up of at least 20 subsamples from points along the ‘W’.

If in doubt, sample any different soil types or areas of the field separately. This might include:

  • An area that was a separate field in the past
  • Part of the field that has had a different crop or fertiliser regime in recent years
  • A different soil type
  • Areas that are particularly low or high yielding
  • Different areas of topography – for example valley bottoms, steep slopes or hills.

Caution in no-till systems

In notill systems there can be less vertical soil movement, particularly in the early days before earthworms and other soil microfauna increase and begin to move the soil.

No Till Farming

This can lead to distinct layers in the soil including:

  • Acidic subsoil where lime has not penetrated and raised the pH
  • Acidic topsoil where acidic ammonium fertilisers have raised the pH
  • Concentration of applied phosphate fertilisers in the top few centimetres. In damp years the crop continues to access the phosphate, but when dry the nutrient is much less accessible to the crop.

A large amount of research has been carried out on how to remedy these effects by deep P placement and regular liming, but the most important factor is to understand what is happening in your soils by testing.

The below graph from North Dakota State University shows particularly well how P stratification can occur in the soil.

P Stratification in the soil

P Stratification in the soil

The suggestion here is that a mix of cover crops (in a corn – soybean rotation) actually increase stratification by bringing P up from depth (thanks to varied and deeper rooting). But the general picture is clear; in long term no till soils the P available in a dry season at depth can be markedly reduced.

Starter P vs Deep P

Zones vs grid sampling

This debate has been raging ever since the introduction of precision farming and is a subject that most agronomists and farmers have a strong opinion on!

Soil Sampling

My belief is that neither approach is too far wrong. Grid samples are arguably easier to administer, whereas zones allow aspects of local knowledge to be brought into the sampling. The important point is to A) do something, B) review the results, and C) to look at where the gaps in knowledge are and sample further if necessary.

zone vs grid sampling

Till next time,


About David


David Jones is the Broad Acre Specialist at Crop Nutrition Laboratory Services Ltd. (CROPNUTS). David has a keen interest in soils and no till farming systems where he has undertaken work looking into weed levels and changes in soil structure, and has extensive experience in field trials and in the development of precision farming techniques. In his spare time he enjoys playing rugby.