A Complete Guide to Variable-Rate Fertilizer Application

I’ve now been carrying out field experiments for 10 years. My working theory behind variable-rate fertilizer application is that high-productivity zones require more fertilizer to be applied, while low-productivity zones require less. Nutrient removal in the high-productivity zones is higher because each yield unit needs a specific amount of nutrients to grow. As such, we should apply the most fertilizer to high-productivity zones to replenish nutrients. In low-productivity zones, the chemical and physical properties of the soil oftentimes don’t affect the yield. Nutrient removal is lower here, which is why these zones require less fertilizer.

During my years of experiments, the working theory saw different results. At times, variable-rate fertilizer application significantly increased yield. Other times, it showed an insignificant increase in yield and huge savings on resources, and still other times, it didn’t impact yield but saved resources. Let’s take a look at how the two winter wheat fields reacted to variable-rate fertilizer application.

Productivity zones are areas of the field that produce the same amount of yield (high, medium, or low) from year to year. If an area in the field produces high yields several seasons in a row, we can consider it to be a high-productivity zone. Medium and low yield zones are medium- and low-productivity zones, respectively.

Why did I choose to apply fertilizers by productivity zone?

• Conducting a soil nutrient analysis isn't enough to determine the optimal amount of fertilizer to apply. The analysis may give you a wrong idea of the soil's condition (read here to see why). Productivity zones give a clear picture of yield based on the satellite images processed with our algorithms. That's one reason I chose this method.

• Moreover, if you choose to apply fertilizers by productivity zone, you also save money. Use a yield monitoring system on your equipment or the OneSoil web app to identify your productivity zones. That's reason #2.

Field #1. We'll call it Jaime.
Locations and soils: Northern Ukraine, sod-podzolic soils. The crop was winter wheat. On September 15, 2020, diammophoska was applied at varying rates: 80, 135, and 190 kg/ha.

Field #2. Let's call this field Cersei.
Locations and soils: Central Ukraine, chernozem soils. On October 1, 2020, diammophoska was applied at varying rates: 80, 135, and 190 kg/ha.

To define the productivity zones in the Jaime Field, we need to first analyze NDVI (normalized difference vegetation index) data.

In the OneSoil web app, I can get this data for four successive years for this field. Then, I find images in the key growth stages when the correlation between yield and NDVI data is most obvious.

NDVI data from year to year shares a lot in common. This points to the following conclusions:

a) The productivity zones are stable.
b) There is a limiting factor in the field that has nothing to do with the crops or the weather. This could be the relief, soil acidity, or any other internal factor.

Combine the images, and you end up with productivity zones.

Green areas indicate high-productivity zones, orange areas are medium-productivity zones, and red areas are the least productive zones in the field.

To avoid misinterpreting the experiment results, we have to find the limiting factor. To do that, we have to compare the productivity zones with the relief and soil brightness.

Combine the productivity zones and the soil brightness map. It turns out that the high-productivity zones are the dark-soil areas rich in humus, while the low-productivity zones are located in light soils.

The relief has no significant impact on the distribution of humus across the field. I hypothesized that the yield in this field is impacted by soil acidity, and the field scouting done in 2021 confirmed this hypothesis.

Now that you know all the productivity zones and the limiting factor, put together a prescription map for equipment. Check out our Guide to learn how to do#nbsp;that in the OneSoil app.

Variable-rate fertilizer application was done in Jamie using the prescription map below:

In the high-productivity zones, the fertilizer application rate was increased to 190 kg/ha, while it was reduced to 85 kg/ha in the low-productivity zones. We left control strips in the high and low-productivity zones to help us interpret the results.

We need control or test strips to confirm the working theory. Using the control strips with the standard rate of fertilizer application, we check whether variable-rate application affects the yields in this specific zone. For example, in the high-productivity zone, we apply 190 kg of fertilizer. In the same productivity zone, we keep a control strip, where we apply 130 kg of fertilizer. It's highly probable that we'll get objective results at the end of the season.

Find more about control strips and other things you should know to conduct the ideal field test in our blog post.

Now back to Jamie. Let's check the field's NDVI data during the season:

The maps show that by mid-season and at the end of the season, NDVI is higher in high-productivity zones that had an application rate of 190 kg/ha. The zones with an application rate of 130 kg/ha don’t develop as actively.

July 25, 2021, was the big day for Jamie: the harvest season began. Combines with yield monitoring systems were used to gather the harvest. After that, we combined the yield map from the combines with the prescription map and got the following results:

The yield map points to the strong impact of fertilizers in high-productivity zones.

Now it’s time to do some manual work. Look at the yield map to find replications, i.e., areas where the yield matches the application rate in the specific productivity zone. To get a reliable result, we have to compare the average yield for the replications in the high-, medium-, and low-productivity zones. Also, for the validity of the experiment, we need to get rid of all the noise in our data.

Find the replications and fill out the data in the chart (see Step 4).

Jaime

In high-productivity zones, the wheat yield increased by 0.53 t after the fertilizer application rate was increased. In low-productivity zones, no impact from the fertilizer application rate on the actual yield was noted.

Cersei

In the low- and medium-productivity zones, no impact from the fertilizer application rate on the yield was noted. In the high-productivity zones, yield increased but insignificantly (0.16 t).

However, to see how well a specific field will react to variable-rate fertilizer application, you need to conduct field experiments in that field’s particular soil and climatic conditions.