Take 15–20 cores in a zig-zag across one uniform area, all to the same depth (about 15–20 cm), skip edges and odd spots, mix in a clean plastic bucket and send about 0.5 kg.
Soil properties vary a lot within a field, from metre to metre. A spot where manure was heaped, where a bund was levelled, where water stands or where ash from burnt residue fell can differ several-fold in phosphorus, potassium, pH or organic carbon from the field average. A laboratory analysis is very precise about the sample it receives; if that sample represents one odd spot, the fertiliser recommendation will be wrong for the whole field. Most of the error in soil testing comes from sampling, not from the laboratory.
Divide the farm into sampling units that are uniform in soil type, slope, cropping history and management, usually no more than a few hectares each. In each unit, walk a zigzag or W pattern and take 15 to 20 small cores or slices with an auger, probe or spade. For a spade, dig a V-shaped hole and take a thin slice about 2 centimetres thick from one side, top to bottom.
Sample to the plough depth, usually 0 to 15 or 0 to 20 centimetres for field crops, and keep every core the same depth; deeper sampling (for example to 60 or 90 centimetres) is used for orchards, for nitrate and for salinity. Avoid field edges, bunds, areas near trees, roads, channels, compost and manure heaps, fertiliser bands, burn patches and wet hollows.
Put all cores in a clean plastic bucket or on a clean sheet. Galvanised (zinc-coated) buckets and brass tools contaminate the sample for zinc and copper; fertiliser bags carry residues. Break up clods, remove stones, roots and debris, and mix thoroughly. Reduce the bulk by quartering: spread the soil, divide it into four, discard two opposite quarters, and repeat until about half a kilogram remains. Dry it in the shade if it is wet; do not dry in the sun or an oven.
Label the bag with your name, field identification, date, depth and the crop you plan to grow next, because recommendations depend on the crop.
Sample after harvest and before fertiliser is applied, and at the same time of year each time, so results can be compared. Soil test every two to three years for most fields, more often for intensive vegetables. Keep the reports together; trends are as informative as any single result.
A standard soil report gives pH, electrical conductivity (salinity), organic carbon, and available nitrogen, phosphorus and potassium, often with sulphur and micronutrients such as zinc, iron, manganese, copper and boron. Each is rated low, medium or high against crop-specific ranges, and the report recommends fertiliser rates for the crop you named. Check that the method used for phosphorus suits your soil (for example Olsen for neutral and alkaline soils, Bray for acid soils), because the numbers from different methods are not comparable. Results that look very different from the previous test usually point to a sampling problem; resample before changing your fertiliser plan.
One handful of soil from one spot is not a soil test. Here is how to take a sample the laboratory can actually use.
Pick one uniform area of your field. Walk across it in a zig-zag, and take fifteen to twenty small cores, every one to the same depth, about fifteen to twenty centimetres.
Skip the edges, the ground near trees, old manure heaps and wet hollows.
Mix all the cores in a clean plastic bucket, not a galvanised one, then send about half a kilogram, labelled with the field and your next crop.
How to read the report is in our long video, and at laisha.ai.
Last updated · How Laisha writes and checks its pages · Report a mistake
US$10 a month · one price worldwide · cancel any time
An AI operating system for agriculture: free guides, manuals and tools written for every country.