Soil testing and fertiliser recommendation
How to sample soil, what common lab tests measure, how results are calibrated into low, medium and high, and targeted-yield fertiliser sums.
On this page
A bag of fertiliser applied without knowing what the soil already holds is a guess. Too little leaves yield in the field; too much wastes money and pollutes water. Soil testing replaces the guess with a measurement: a sample from the field is analysed for the nutrients plants can use, and the result is turned into a fertiliser recommendation for a particular crop and yield.
Soil testing is used worldwide, from national soil health card schemes for smallholders to precision grid sampling on large farms, and the chain from sampling to recommendation is a standard exam topic in soil science and agronomy.
The chain from field to recommendation
A soil test is only as good as each step:
- Sampling a representative composite sample.
- Analysis with a method suited to the soil.
- Interpretation: deciding whether the value is low, medium or high for the crop.
- Recommendation: how much of which nutrient to apply, when and how.
Errors at the first step are the most common and the hardest to detect, because the laboratory can only analyse what it is sent.
Sampling
- Divide the farm into uniform units by soil type, slope, colour, crop history and past management; sample each separately.
- In each unit, take about 15 to 20 cores in a zigzag or grid pattern, avoiding field edges, paths, manure heaps, fertiliser bands, and spots under trees.
- Sample to the plough layer, usually 0 to 15 or 0 to 20 cm for field crops; deeper layers for mobile nutrients such as nitrate and for tree crops.
- Mix the cores thoroughly in a clean bucket, reduce to about 500 g by quartering, air-dry in the shade, and label with field, date and depth.
- Sample at the same time of year each time, before sowing or fertilising, and repeat every two to four years for immobile nutrients.
Precision agriculture uses grid sampling (for example one composite per hectare) or zone sampling based on yield maps and soil sensors, so that fertiliser can be varied within the field.
What the laboratory measures
Laboratories measure an index of the plant-available fraction, not the total nutrient. Different extractants suit different soils, so results are only comparable between labs using the same method.
| Property | Common methods |
|---|---|
| pH and electrical conductivity | 1:2 or 1:2.5 soil–water suspension, or in calcium chloride |
| Organic carbon | Walkley–Black wet oxidation or dry combustion |
| Available nitrogen | organic carbon or mineralisable N as an index; soil nitrate tests (for example a pre-sidedress nitrate test in maize); alkaline permanganate in some countries |
| Available phosphorus | Olsen (sodium bicarbonate, pH 8.5) for neutral and alkaline soils; Bray-1 for acid soils; Mehlich-3 for a wide range of soils |
| Exchangeable potassium | neutral normal ammonium acetate, or Mehlich-3 |
| Micronutrients (zinc, iron, manganese, copper) | DTPA extraction (Lindsay and Norvell, 1978); hot water for boron |
| Sulphur | calcium chloride or phosphate extractable sulphate |
From soil test value to fertiliser rate
Correlation and calibration
A test is useful only if its values relate to crop response. Correlation studies check that the test value tracks nutrient uptake or yield across many soils. Calibration trials, run over many sites and seasons, measure how much yield increases when the nutrient is applied at different soil test levels.
The critical level is the soil test value above which a response is unlikely, often taken as the value giving about 90 to 95 per cent relative yield. The Cate–Nelson method, which splits the data into responsive and non-responsive groups, is a classic way to find it. Results are then grouped into ratings such as low, medium and high (sometimes very low and very high too), each with a recommended rate.
Recommendation philosophies
- Sufficiency level of available nutrient: apply fertiliser only when the test is below the critical level, at a rate that rises the lower the test. It maximises short-term return.
- Build-up and maintenance: raise a low soil to the critical level over a few years, then apply what the crop removes to keep it there. It costs more at first but simplifies later decisions.
- Basic cation saturation ratio: aims for "ideal" ratios of calcium, magnesium and potassium; research has generally not supported fixed ideal ratios, and it is used less today.
- Targeted yield: calculates the fertiliser needed for a specific yield from the crop's nutrient requirement and the contributions of soil and fertiliser.
Targeted-yield nitrogen recommendation
Field trials for a wheat variety give:
- Nutrient requirement (NR): 25 kg N per tonne of grain.
- Contribution from soil (CS): 20 per cent of the soil test N.
- Contribution from fertiliser (CF): 50 per cent of applied N.
The targeted-yield equation is: fertiliser N = (NR ⁄ CF) × target yield − (CS ⁄ CF) × soil test N, with CS and CF as fractions:
- fertiliser N = (25 ⁄ 0.50) × T − (0.20 ⁄ 0.50) × SN = 50 T − 0.4 SN.
- For a target of 5 t/ha and a soil test N of 250 kg/ha: 50 × 5 − 0.4 × 250 = 250 − 100 = 150 kg N/ha.
- Using urea (46 per cent N): 150 ÷ 0.46 ≈ 326 kg urea per hectare, usually split between sowing and early growth.
The coefficients are illustrative; real ones come from local soil test–crop response trials and differ by crop, soil and region.
Converting a soil test to kilograms per hectare
A soil has an Olsen P of 8 mg/kg in the top 15 cm. With a bulk density of about 1.33 g/cm3, a hectare of soil 15 cm deep weighs about 10,000 m2 × 0.15 m × 1,330 kg/m3 ≈ 2,000,000 kg.
- Olsen-extractable P = 8 mg/kg × 2,000,000 kg ≈ 16 kg P per hectare.
- Suppose the critical level is 16 mg/kg, and local data show that about 15 kg P/ha raises Olsen P by 1 mg/kg. Building up from 8 to 16 mg/kg needs about 8 × 15 = 120 kg P/ha, or 120 × 2.29 ≈ 275 kg P2O5/ha, usually spread over several seasons on top of crop removal.
The "buffer" figure of 15 kg P per mg/kg is soil-specific: strongly fixing soils need much more.
Plant analysis
Leaf or tissue analysis complements soil testing, especially for tree crops and micronutrients. Values are compared with critical concentrations or with ratios between nutrients, as in the DRIS (Diagnosis and Recommendation Integrated System). Plant analysis shows what the crop actually took up, but often too late to correct the current crop.
Key points
- Soil testing chain: sampling, analysis, interpretation, recommendation; sampling causes most errors.
- Take 15–20 cores per uniform unit to plough depth, mix, quarter, air-dry and label.
- Methods index available nutrients: Olsen P for neutral and alkaline soils, Bray-1 for acid soils, Mehlich-3, ammonium acetate K, DTPA micronutrients.
- Calibration links test values to yield response; the critical level marks where response becomes unlikely.
- Philosophies: sufficiency, build-up and maintenance, cation ratios (little support), targeted yield.
- Targeted yield: fertiliser N = (NR ⁄ CF) × T − (CS ⁄ CF) × soil N.
What the exams ask
- School-leaving and entrance papers: soil sampling procedure, the purpose of soil testing, and soil health cards or reports.
- Degree and research-fellowship entrances: extractants and methods for each nutrient, correlation and calibration, critical levels and Cate–Nelson, recommendation philosophies, and targeted-yield equations.
- Professional examinations: interpreting soil reports, calculating fertiliser and product rates, precision sampling, and plant analysis with DRIS.
Described in general terms: the kind of examination, not a past paper. Laisha never reproduces exam questions.
Questions people ask
How often should I test my soil?
Every two to four years for phosphorus, potassium, pH and organic matter, at the same time of year. Mobile nutrients such as nitrate change quickly and are tested within the season when needed.
Why do two labs give different numbers for the same soil?
Different extractants dissolve different fractions of a nutrient, so an Olsen P and a Bray P of the same soil differ. Compare results only with the ratings for the method used.
Is a soil test enough to set fertiliser rates?
It is the foundation, but rates should also allow for the crop, yield target, previous crop, manure applied and local trial data.
Sources
- Brady, N. C. and Weil, R. R. The Nature and Properties of Soils, 15th edition. Pearson, 2017. Chapter 16.
- Havlin, J. L., Tisdale, S. L., Nelson, W. L. and Beaton, J. D. Soil Fertility and Fertilizers, 8th edition. Pearson, 2014.
- Olsen, S. R., Cole, C. V., Watanabe, F. S. and Dean, L. A. Estimation of Available Phosphorus in Soils by Extraction with Sodium Bicarbonate. USDA Circular 939, 1954.
- Lindsay, W. L. and Norvell, W. A. "Development of a DTPA soil test for zinc, iron, manganese, and copper." Soil Science Society of America Journal 42, 421–428 (1978).
- Cate, R. B. and Nelson, L. A. "A simple statistical procedure for partitioning soil test correlation data into two classes." Soil Science Society of America Proceedings 35, 658–660 (1971).
More in soil science
Discussion
Questions, corrections and notes from readers on this article. Every comment is checked before it appears: keep it on the topic, and do not share phone numbers, links or contact details.