Time to Take Soil Samples: Some Things to Consider

by Leo Espinoza, Extension Soil Scientist - October 13, 2021

Fertilizer prices are significantly higher now than last year, and farmers know that soil sampling is one of the tools available to optimize fertilizer use. In Arkansas, about 65-70% of the soil samples submitted to the U of A, Marianna Soil Testing and Research Laboratory are collected for Variable-Rate Fertilization (VRF) purposes.

A condition for successful VRF is developing prescription maps representing the true variability of the nutrient of interest. What factors affect the quality of a prescription map?

Some factors include sampling depth, collection technique, sampling strategy, sampling density, sampling time, and interpolation method. Recommended sampling depths are 4 inches for rice and soybean and 6 inches for corn and cotton. This is because fertilizer recommendations were developed using samples collected at the corresponding depths.   Soil samples shallower that the recommended depth will tend to overestimate the concentration of the nutrients due to stratification.

Then, maybe the most important question remains: what is the best way to collect soil samples?   Are grids better than management zones? Nutrient spatial variability is a result of the complex interactions between soil, climate, and management history. Therefore, soil fertility can vary greatly, even within short distances. The idea that any sampling strategy for VRF is suitable because you collect more than the traditional 1 sample every 20 acres is often incorrect and may give a false sense of precision. Below are some thoughts for your consideration.

1. There isn't a sampling approach that fits all soils and situations.
2. The sampling strategy that best describes the variability of a particular nutrient may not be the same for another nutrient. For instance, it could be 3-acre grids for phosphorus and 5-acre grids for potassium.
3. You may want to prioritize potassium over phosphorus.
4. The optimum sampling resolution may vary even within a single cropping system.
5. While the emphasis has been on accounting for nutrient variability across a field (spatial variability), the variability across time (temporal variability) can be as high. The behavior of nutrients in the soil is affected by temperature and moisture, and thus, nutrient levels will tend to be higher under warm than cold weather. So, be as consistent as possible with sampling time.
6. If you follow a sound nutrient management plan, soil test values should not drop significantly after a couple of growing seasons.
7. Fields with random or short-distance variability (microtopography, sand blows, etc.) can be challenging to manage with VRF.
8. If the variability in a field is "random," a small grid size (less than 3 acres) is probably the best approach. Management zones or larger grid sizes are probably more appropriate for structured variability, such as land leveling and varying soil series or texture.
9. A soil test is an index of nutrient availability, and if the same sample is analyzed several times, results can vary by as much as 10%. Repeated analysis of samples testing "low" could vary by more than 10%.
10. Avoid taking soil samples "behind" the combine. Crops like corn need about 10 inches of rain to release the potassium stored in the stubble.

Producers need to be aware that different interpolation methods can be used to generate prescription maps from a set of data points. An interpolation method is a procedure to estimate nutrient concentrations in areas that were not sampled. The choice of interpolation method is crucial as each method may result in different amounts of fertilizer being recommended for the same field.

Finally, it is not realistic to expect that all the nutrient variability in a field can be managed with VRF. Studies have shown that fertilizer spreaders can show variations of up to 15% of the target rate even if adequately calibrated, and may need 6 seconds (40-60) feet to change fertilizer rate.