Yield maps have been known in agricultural practice for a good 25 years. The measurement of sub-field-specific yields and the resulting production of yield maps for a field have improved enormously during this time. Today, their quality can therefore be considered quite good. The question now arises: what do you do with yield maps? There are two possibilities:

                              (1) Weak point analysis of the production result of one year, good idea!

                              (2) Basis for deriving crop management decisions for the following year, better not!

Why not?

Yield potential maps are to be calculated from a series of yield maps. This, in turn, is to be used in precision farming to be able to plan management measures and the use of inputs. At first glance, this idea seems to make sense. This is why many farmers are also concerned with the topic and have invested in yield measurement systems in combine harvesters, for example.

However, the wish is father to the thought! I can still remember the time around the turn of the millennium when many users had multi-year yield maps available for the first time. The sobering conclusion was: "the more yield maps we collect, the less we understand." 

Excuse me? Is this supposed to mean that the more concrete the data, the more clearly it becomes apparent that yield potential maps do not work?

We have to look more closely!

Prof. Simon Blackmore (Harper Adams University) already investigated the possibility of predicting future yields in 2002 using the structure of 6-year yield maps from Denmark. The aim was to create yield potential maps in order to be able to derive future management measures. Prof. Blackmore came to five key conclusions:

1. The yield differences of the individual years can have the greatest influence on the yield.
2. The spatial variability (yield pattern) of the individual year is strongly pronounced.
3. The yield patterns of several years cancel each other out.
4. Yield trend maps cannot predict the yield of the following year.
5. Therefore, instead of the yield, the current demand must be managed.

Yield level, yield pattern and predictability of a future yield

Due to Prof. Blackmore's exact scientific evidence, we repeated the analysis with yield maps from Germany. We examined 5- to 8-year yield maps from a total of seven fields from Mecklenburg-Western Pomerania, Saxony-Anhalt, Thuringia and Saxony. The precise question was whether one can forecast future yields with sufficient certainty on the basis of historical yield data? Our results were amazingly clear!
However, one must first distinguish between two effects in the data analysis:

(A) Effect no. 1:

➔ Annual effect of yield formation or where is the average yield of the field?
For each field, one can calculate an average yield over a period of time. In our example data set, these average yields ranged from 54 to 89 dt/ha, depending on the field and the number of years considered.

➔ But the exciting question is now, how far can the individual years deviate from the long-term average yield of the field? In our sample, the individual year deviates on average by around 17 dt/ha (7 to 28 dt(ha) upwards and downwards across all fields. This means that the year effect alone (good and bad years) leads to a yield variability of the average yields of fields in the order of about 34 dt/ha. Who would dare to predict the average yield of a field in October or March with a quality of about 5 dt/ha? That person would be a made man at the hail insurance company or at the brokers!

A crucial question remains open

Can we predict future single-year yields from multi-year yield maps? We have tried to do so using this example data set. We used the prediction quality (R²) as a yardstick. This is 0.15 on average, which means that only 15% of the yield of a single year can be explained by the historical data. The best values were 41%, the worst 0%. In other words, the historical data cannot be used to draw conclusions about the yield of the coming year.

Now, one can argue that historical yield data alone do not lead to a yield potential map. One should also include soil samples and weather data, among other things. I have also been hearing this argument for a good 25 years now. But in that time, I have not found anyone who has been able to show that this works anywhere. After all, if you know that, for example, soil type correlates positively as well as negatively with yield, or that no one can predict a dry year or a wet year, or a cold year or a warm year, then this approach fails for practical purposes.

What one does find in the relevant literature, however, are scientific studies that come to more or less the same conclusions, namely the unpredictability of future yields.

1. The future yield cannot be predicted with sufficient certainty from the observation of multi-year yield maps for the derivation of operational crop management measures. The agronomic approach of planning input costs on the basis of yield potential maps cannot be supported (or only to a limited extent).
2. One should focus on managing the current variability within a year. It is important to recognise limiting factors during the growth processes that limit the possible yield of a year (water supply, temperature, global radiation) and to determine their effects (current N requirement, infection pressure, occurrence of weeds, etc.) on the input use.
3. Instead of the yield target, the current demand must be managed.
4. This analysis does not question the general use of yield mapping systems. They are still an excellent tool for the annual analysis of weak points, among other things. However, the data generated with them cannot fulfil the function they were intended to perform.

It never ceases to amaze me that the topic of the yield potential map keeps finding its way into agricultural magazines and at agricultural conferences. In my opinion, its proponents have never had to put their claims to the test. It is and remains mere theory - without feasibility and accuracy. The fact that it is nevertheless heard in practice is astonishing and sobering. Nor do I understand why yield potential maps are regularly mentioned in connection with plant sensors (not the YARA N-Sensor®!). Because I observe with concern that individual scientists, advisors and also sellers of these sensors always dump the task of creating alleged yield potential maps on the farmer. And if the calculation does not work out, it is not the plant sensor's fault, of course, but the farmer's. He simply does not have the right yield potential. The farmer simply did not produce a correct yield potential map! This is how "easy" it can be to keep a pipe dream alive.