Latest articles

Soil sampling and soil testing

Fall is generally the most convenient time for soil testing. Plan now for effective testing

In the last issue of Grainews Les Henry wrote a great article on soil testing. I couldn’t agree more that soil sampling and testing is a great tool to assess the soil nutrient levels in your fields so you can wisely plan your fertilizer program for next spring. Here are some of my thoughts on soil sampling and testing.

When to soil sample?

Soil sampling annually-cropped fields just before spring seeding gives the most accurate measurement of soil nutrient status. But realistically, spring is often too short and rushed to allow soil sampling, analysis and developing your fertilizer plans. So, sampling in late fall after soil temperature has dropped to 5 to 7 C is often the most practical time.

But, it is important to remember that plant available soil nutrients like nitrogen, phosphorus and sulphur will fluctuate from fall to spring, particularly if soils are moist with warmer-than-normal conditions. Variation from fall to spring is greatest in the Chinook regions of the southern Prairies. There are years when we have to compensate for over-winter changes of soil nutrients. Also, I am not a fan of sampling frozen soils simply because of the difficulty in obtaining representative sampling depths.

Sampling options

Many fields across the Prairies have moderately rolling topography resulting in considerable soil variability. This is a major challenge when deciding how to take representative soil samples. Samples must be representative of the field or each management zone of a field. Work with your fertilizer dealer or agronomist to decide how to sample each field. I strongly suggest you go out with the person doing the soil sampling on your farm to ensure sampling is done in appropriate areas in your fields and to make sure enough sites are sampled in each field. When you are with the sampler you know where and how the samples were taken.

There are a number of ways field soil sampling can be done. The three more common sampling methods are:

1. Random sampling of a whole field works best in fields with relatively uniform soil and topography. It involves taking representative samples throughout the entire field, but making sure to avoid unusual areas.

2. Sampling soil/crop management zones works best in fields with variable soil and topography. Uniquely different zones are mapped based on soil characteristics, topography, and/or crop yield potential. Representative soil samples are taken within each zone. This works well in fields with variable soil. Each management zone can be randomly sampled or benchmark sampled. Consider working with an agronomist to prepare zone maps for each field.

3. Benchmark soil sampling involves sampling a one to two acre area representative of the majority of the field or zone. Sample the same area each year. When a field is variable in soil or topography, three or more benchmark locations may be needed to represent the different field areas.

When selecting soil/crop management zones make use of crop yield maps, aerial photos, topographic maps, soil salinity maps and satellite imagery information. Also, use your personal field knowledge and observations of crop growth differences (crop establishment, vigor, colour and growth) and topography of each field to identify where different soil types occur.

How many sampling sites?

I suggest taking a minimum of 20 soil sampling sites for each field, zone or benchmark area. Les Henry suggests 30! The more sampling sites, the more representative your samples will be of the field. A common mistake is only taking six or seven soil cores from a field or management zone, which is not enough and can result in unreliable information and inaccurate fertilizer recommendations.

Why do we need so many samples? Typically, each soil sample sent to a soil testing lab weighs about two pounds. One acre of land, six inches deep, weighs about 2,000,000 lbs. If a 160 acre field is soil sampled to six inches, a two pound soil sample must be representative of 320 million pounds of soil. The soil sample would represent about 0.0000006 per cent of the field. This is an extremely small representation of the total field. Therefore, it is very important that an adequate number of soil cores be taken!

What depth increments to sample!

Various sampling depths are suggested but my preference is to separate each soil core into depth intervals of zero to six, six to 12 and 12 to 24 inches. Place the three sampling depths into three clean plastic pails. Do this at each site sampled.

Most soil test correlation research with annual crops in Western Canada with nitrate-nitrogen and sulphate-sulphur has been with sampling to 24 inches. But, it is also important to know the distribution of N and S in the soil profile. Sampling three depths will give a good picture of the amounts of each nutrient and where the nutrients are located in the soil profile. For P and K, most soil test correlation research has been with the surface zero to six inch depth, as these nutrients are less mobile in soil than N and S.

Some agronomists and dealers prefer to only sample one or two depths making the sampling process simple and faster. If only the zero to six and six to12 inch depths are sampled, you have no idea of the amount of N or S that may be present in the subsoil — this information is important to develop accurate N and S fertilizer recommendations. If the zero to six and six to 24 inch depths are sampled, the results of an extended sampling depth of 18 inches (six to 24 inch depth) can be more difficult to interpret and sometimes misleading.

After taking the 20+ soil cores, thoroughly mix each composite sample and lay out the soil samples to completely air dry to stop nutrient changes. If moist soil samples are sent to the lab in sealed bags, soil microbes can alter the levels of plant available nitrogen, phosphorus and sulphur causing incorrect estimates of soil nutrient levels. To air dry, spread the soil sample in a thin layer onto clean paper, plastic sheets or place into clean, shallow plastic or aluminum trays. Dry the samples at room temperature in a clean room (no cats or other animals, to prevent contamination). Do not use artificial heat to dry samples. If moist samples are sent directly to the lab, they must be shipped in coolers, kept below 5 C and arrive at the lab the next day for drying

What analysis is required?

The important plant-available macronutrients to test for are: nitrate-nitrogen (NO3N), phosphate-phosphorus (PO4-P), potassium (K+), and sulphate-sulphur (SO4-2-S). Determine plant available N, P, K and S in the zero to six and six to 12 inch depths and test for N and S in the 12 to 24 inch depth. Normally, there is no need to test for plant available calcium (Ca+2) or magnesium (Mg+2) as these nutrients are very rarely deficient in Western Canada. It is a good idea to occasionally check the soil micronutrients copper (Cu), iron (Fe), manganese (Mn), zinc (Zn), boron (B) and chloride (Cl). Testing for micronutrients every year is only necessary if one or more micronutrients are in the marginal or low range; otherwise testing every four years is fine. It is important to realize the tests for B and Cl are not very reliable. Often soil analysis levels are interpreted as low for B or Cl, but crops do not respond to added fertilizer. The problem is with the soil test methodology and critical levels used, resulting in unnecessary fertilizer recommendations.

Determining soil organic matter, pH (a measure of soil acidity/alkalinity) and electrical conductivity (E.C. — a measure of salinity) are useful to monitor soil chemical properties of your fields. Some agronomists may recommend determining Cation Exchange Capacity (CEC) and determining base cation saturation ratios. Research has shown this is not a useful determination for making fertilizer recommendations for most soils or crops in Western Canada.

Finally, make sure the soil testing lab uses the correct soil test methods. For Alberta farmers, all soil test P calibration has been with the Modified Kelowna method since 1990, Alberta samples should be sent to a lab that uses this method. For Manitoba farmers, all soil test P calibration has been with the Olsen method (also referred to as the Bicarb method). For Manitoba farmers, use a lab the uses the Olsen method. For Saskatchewan farmers, either method can be used to determine soil P. Other soil test P methods, such as the Bray method, have never been calibrated to Western Canada soils. It is my opinion that other methods that have not been calibrated for western Canadian soils should not be used.

Most labs report N, P, K and S levels in pounds per acre (lb./ac.) and the micronutrient levels are reported in parts per million (ppm). The process of soil test interpretation is the next step in the process. Make sure you seek the advice of several agronomists when developing your fertilizer plans for next spring.

About the author


Ross H. McKenzie, PhD, P. Ag., is a former agronomy research scientist. He conducted soil and crop research with Alberta Agriculture for 38 years. He has also been an adjunct professor at the University of Lethbridge since 1993, teaching four-year soil management and irrigation science courses.


Stories from our other publications