How to Increase Soil Testing Quality

Soil testing is a study of soil conducted to identify its content and to evaluate the availability of nutrients for plants. Typically, soil testing is performed every three to five years. Soil samples are taken in the field, sent to a laboratory, and analyzed.

More than 40 soil properties can be determined during soil testing. The most common indicators are the content of organic matter (humus), the level of soil acidity, and the availability of phosphorus, potassium, nitrogen, and other elements.

In this article, we want to explain why the most rational approach in soil testing is sampling by field productivity zones.

Productivity zones are areas of a field with different yields. A high productivity zone is an area that has consistently yielded the highest harvests over several seasons. Conversely, a low productivity zone continually produces low yields, while an average productivity zone gives average yields.

The primary purpose of defining productivity zones is to forecast yield distribution for the upcoming season. The higher the quality of the productivity maps, the more accurately yields can be predicted.

At OneSoil Yield, we create productivity zone maps using our unique algorithm. This method enables us to build incredibly accurate productivity maps, the most precise we've compared thus far.

The techniques for sampling and methods of soil testing can differ widely, depending on the country or region. Many organizations partition the field into a grid composed of square or rectangular plots. From each plot, they collect 10-20 soil samples to create a composite sample. In some cases, soil samples may be gathered from fixed points within the field, after which soil property values are interpolated.

We won't get too deep into discussing which methods are optimal for examining nutrient content; we trust that you'll know the best approach for your soil.

Let's take a look at a field in central Ukraine. The soils are chernozem, characterized by a loamy granulometric composition. In the OneSoil Yield application, the field's productivity zones are displayed as follows:

Our algorithm has generated a field productivity map based on five out of the available seven seasons. In other words, during five of these seven seasons, vegetation in the field followed the same distribution pattern. This implies that the high and low productivity zones remain consistent year after year.

By utilizing productivity zones, we can anticipate yield distribution for the coming season: the "red" zones are likely to yield less, while the "green" zones will probably yield more.

We can even attempt to understand why productivity is divided into such zones. Let's take a closer look at the topography and soil brightness (which indicates the distribution of humus):

As observed, the high productivity zone is situated in areas with darker soil, which suggests higher organic matter and moisture content, and is predominantly in lower-lying areas. Conversely, the low productivity zone is located on elevated and sloped lands with lighter soil, indicating lower organic matter and moisture content. This indicates that soil erosion may be a problem in low productivity zones, with low moisture reserves further limiting yields in these areas.

We have yield data for this field from 2021. To evaluate the accuracy of our productivity zones, let's compare them with this yield data:

Productivity zones correlate well with the yield data. This confirms the quality of the OneSoil yield zones.

To evaluate the best way to choose zones in the field for soil testing, we will use the yield map as our guide.

We divide the field into a grid of plots. On the left, you see a regular grid created without considering productivity zones. On the right, you can see the plots designed for soil testing, based on productivity zones.

In both maps, the field was divided into 23 plots, each averaging 4.6 hectares in size.

Now, let's superimpose our selected zones onto the yield map and observe how the data we collect from the soil tests correlate with the yield data.

To obtain a soil sample, we conducted 15 soil samplings from different locations within the same plot and sent them to the lab. We received nutrient content values for the area within the plot boundaries and tried to draw conclusions from them.

Because you may encounter very different soils with varying potential fertility within the same plot, the data you obtain might be unrepresentative. For "red" grid plots, yields can vary from 7 t/ha to 13 t/ha. That's nearly a two-fold difference.

In summary, you will receive very inaccurate results for 10 out of the 23 plots.

This data cannot be used to understand how field yields relate to humus, phosphorus, and other elements. Consequently, it's impossible to correctly interpret the situation in the field and plan steps to improve soil quality based on this data. In other words, half of your investment in soil testing could be wasted.

If we also conduct 15 soil samplings within the boundaries of the designated plots by productivity zones, the samples will be much more consistent with each other within the same plot.

There are no plots where the yields exhibit significant variations. The data for half of the plots were accurate, and for the other half, extremely precise.

With this map, it's much easier to comprehend how the content of various elements in the soil correlates with yield. For instance, it may show where lower yields correlate with a phosphorus deficiency.

This approach simplifies and enhances the accuracy of planning future operations to improve the soil — for the same cost and with the same number of soil samples.

Bonus: Let's consider the soil brightness map, also available in the OneSoil Yield app. Soil brightness strongly correlates with soil humus content. Dark soil signifies high humus content, while light soil indicates low humus content.

If you take samples on a regular grid, they will reveal that the humus content varies significantly across many areas.

If you utilize productivity zones, you obtain a fairly accurate representation of which areas have more humus and which have less. Moreover, humus content is the primary agrochemical factor that impacts soil productivity.

The key takeaway: using productivity zones substantially enhances the quality of soil testing. With the same number of samples, you'll gain a much more comprehensive understanding of the soil than if you sample within plots using a regular grid.