We would like to cultivate a culture of discussion

...with our agriconBLOG

agriconBLOG

19.12.2018 - Peer Leithold (Send email to Peer Leithold)

Yield potential maps a second time - Why, agronomically, they make no sense at all

In the last blog post on the subject of yield maps, I pointed out that future yields cannot be planned and that it is more important to correctly measure the demand for a resource. Yield potential maps and yield targets are mostly used to measure the optimal nitrogen fertilization. However with this approach, as I will explain hereafter, you are caught on the wrong track. An attempt is made to calculate the required N fertiliser by balancing plant growth, N consumption and N fertilizer input as in a static black box. However, nature is not that simple.

The (non-) correlation between optimal yield and optimal N-fertilization

The level of optimal N fertilization is defined at the point where the costs of the last kilogram of nitrogen are still financially covered by the increase in yield. If this question is to be approached in a scientifically correct way, it is customary to carry out N enhancement trials on plots. The result shows the following example of a 20-year series of experiments in Saxony, representative for many similar experiments in Germany and

worldwide. Winter wheat was always cultivated on the same field. In 13 of 20 years the fertilization optimum is between 25 and 100 kg N/ha above or below the mean value. If one estimates this
deviation cautiously on the basis of an average production function for nitrogen financially, then a yield or nitrogen advantage for the differentiated N fertilization of about 80 €/ha and year results. This is an order of magnitude that is worth considering. Furthermore, the positive effects in terms of storage
avoidance, quality improvement, combine harvest and risk of disease are not included.

In the following graphic the table columns "optimal N-expenditure" and "optimal yield" are shown. It is clearly evident that, based on optimal yields, which, however, cannot be planned (see last post in the blog) I will gladly repeat this in more detail: It is neither possible to reliably forecast the target yield
(in the sense of optimal yield), nor can one directly conclude from this to optimal N fertilization. So here two things come together which exclude this way.

For an optimum yield of approx. 80 dt/ha, sometimes 60, 80, 120, 140, 170, 180, 200 or 220 kg N/ha are required. It should therefore be noted that there is no fixed relationship over time between optimum yield and the amount of N fertilizer required. Only 14% (R²=0.1438) of the amount of optimum N fertilization is explained by the optimum yield. And this is now also the second reason why yield

potential maps do not work when fertilizing. Once again as a reminder:

The future yield cannot be predicted on the basis of historical yield maps or yield potential maps.
There is only a slight correlation between the optimum yield and the corresponding N-fertilization.

Why is that?

The theoretically achievable yield of a sub-area and thus the N consumption of the plants depends essentially on three basic factors:

the water supply,
from the temperature sum and
of global radiation.

These three fundamental factors are exclusively weather-related and therefore beyond our control. In order to achieve this theoretical yield, we have to feed the plants properly, keep them healthy and ward off damage from insects. The N consumption can first of all be covered by the N supply of the soil. In addition to the available soil nitrogen at the beginning of the growing season, mineralisation plays a decisive role. Agriculturally used soils in our latitudes have between 6,000 and 12,000 kg N/ha in organically bound form. Depending on weather conditions and the activity of the microorganisms, between 0.5 and 3% of this is released. This means that the soil could provide a minimum of 30 or a maximum of over 300 kg N/ha for the plants. These release rates are incalculable. A further influencing
factor is the different development of the root system. This is the ability of the plants to actively absorb nitrogen. The consumption of N by plant biomass growth, the provision of soil-borne nitrogen and the ability to acquire it via the root system means that the optimal N fertilization cannot be quantified using
calculation models. The simple way out is to measure the N requirement directly in the plants. Ultimately, the aim is to provide the plants with an optimum supply of nitrogen. So it is obvious to "ask" the plants directly.

What do you have to "ask" the plants? What must be measured?

The first thing we have to figure out is whether there is a:

measurable,
objective,
high-resolution,
easy to handle and
low-cost parameter, which is closely related to the optimal N fertilization.

The table lists the results of various possibilities. The coefficient of determination (R²) indicates how close or how good the relationship of the parameter is to the level of optimum N fertilization. If the R²=1, then we are dealing with a 100% correlation. The smaller the value, the weaker the correlation and the more unsuitable the parameter.

What is the N uptake of a plant stand?

Of all measurable parameters, the N uptake is the one that correlates most closely with the optimum N fertilization. What is the N uptake? The N uptake is the amount of nitrogen contained in one square meter or one hectare of above-ground plant mass. At the moment there is only one measuring tool that has an absolute calibration for the N uptake and that is the YARA N-Sensor®. Even if other suppliers advertise with similar statements, to measure as well as the N-Sensor®, in the end it is untrue. The easiest way to check this is to make sure that the measurement of another method shows the "N-uptake in kg N/ha" in the display or in the documentation.

How do you get from the N uptake to the optimal N fertilization?

The relationship between the current N uptake and the correct level of N fertilisation is different for each crop and EC stage. This means that appropriate perennial plot trials must be set up for all situations and this relationship must be worked out meticulously. An example of the 3rd N application in winter
cereals is shown in the following diagram. These scientific relationships are the basis for the so-called control functions, which are integrated in the YARA N-Sensor®. The control function describes by how many kilograms the N fertilization must be increased or decreased if the N uptake changes by one kilogram. These algorithms are, in addition to the reliable measurement of the N uptake, indispensable for making agronomically correct decisions. Control functions are currently available for all fertilization dates for winter and summer cereals, rape, maize, potato and sugar beet. Only if the control functions are used
correctly can it be ensured that all sub-areas receive the optimum amount of nitrogen. There are basically two types of control functions:

King John (take from the poor and give to the rich)
Robin Hood (take from the rich and give to the poor)

I personally consider the fact that a user is supposed to determine the control functions himself to be a marketing gag. This is rather to cover up the lack of agronomic calibration. In the field, I have often experienced that you quickly fall into chaos when "creating" control functions oneself. The "two-point
calibration" often found in the field suggests that users have everything in their hands. This is not entirely wrong, but the probability of getting it right is negligible. Despite many years of experience in plant nutrition, I do not trust myself to decide whether I should react to the variability of 1 kg N uptake
with 0.5 or 1, 2 or 3 kg N/ha more or less N fertilizer. The result is that (as often experienced in the field with this kind of calibration) the optimal N amounts are missed with 10-40 kg N/ha per fertilization time. If you chooses the wrong control function out of ignorance then one really causes "damage" and it
would be better to return to constant application. This is also the reason why the public often says "Well... the effects do not occur like this". From my experience this is almost always just a question of the correct agronomic control functions.

What about the field?

As explained at the beginning, there are immense differences in the amount of optimal N fertilization from year to year. The same is true within a field in one year. This is most visible to the farmer by means of the yield maps. The yield of a partial area fluctuates by plus or minus 50% around the average value. In other words, we have an infinite number of effort-yield reactions on a field. The challenge is to fertilise them skillfully and precisely. And this is exactly where the positive effects of the precision fertilisation method come into play. On the field, the N uptake or the necessary N fertilisation varies, depending on the situation, in the order of 50-80 kg N/ha on average. Each sub-area has a different N-optimum. This means that some sub-areas require a fertilisation of between 0 and 30, some sub-areas require a fertilization level of 60-100 kg N/ha. In the case of rapeseed it goes even further due to its high appropriation capacity. Measurements of N uptake on uniformly cultivated fields show that this value changes over a very small area. The increases and decreases build up and decrease over a distance of 4, 6, 10 and 15 metres. This high spatial resolution can only be achieved with ground-based sensors, which produce over 100 readings per hectare.