How the Düngeverordnung changed nitrogen decisions at field level
Agricultural sustainability resists a single measurement because it is environmental, economic and social at once. A narrower question can be tested: comparing modelled crop nitrogen demand against satellite-estimated uptake flags likely over-fertilisation at postcode scale across German winter wheat. The output is a research instrument for an operator's own use, not a rating of individual farms. MELODY, started in May 2026, extends the approach toward a composite index across regions in India and Germany.
Where this data comes from
This is our own field research rather than a literature review, which is worth stating because it explains both its value and its limits.
Over a five-year study conducted under a European Space Agency project, we worked with around sixty German farmers on nitrogen recommendations for winter wheat, recording application rates, timing and outcomes. That gives a picture of what was actually applied on specific fields, which national statistics do not.
It is also a sample of sixty farms, self-selected by willingness to participate in a satellite agronomy project. They are probably more technically engaged than the national average. Read the trend rather than the absolute values.
What the numbers show
Two decades, two distinct phases.
From 2000 to around 2010, application on the farms we worked with fell from roughly 220 to about 195 kilograms of nitrogen per hectare. This was technology-led rather than regulation-led, portable sensors, soil testing, better spreader calibration. Nobody was being made to do anything.
From 2017 the decline steepened, reaching roughly 165 kg/ha by 2020 and about 150 kg/ha in 2022. That period covers both the fertiliser regulation and, in 2022, an extraordinary spike in fertiliser prices. Separating the two is not possible from this data and we will not pretend otherwise, they pushed in the same direction at the same time.
A note on the different percentages in circulation
Reconciling this data with published figures requires care, because several reduction percentages get quoted and they measure different things.
Our field study shows roughly a 32% reduction in winter wheat application rate between 2000 and 2022, on the farms we monitored.
Published international comparisons put German wheat nitrogen application around 165 kg/ha in 2000 and about 150 kg/ha by 2018 - a much smaller decline, from a lower baseline. Our figures run higher throughout, which is what you would expect from farms optimising for yield and protein content.
National fertiliser sales fell sharply in 2022 on price. That is a sales volume figure, not a per-hectare application rate on a specific crop, and it should not be compared directly with either of the above.
If you take one thing from this section, take that: nitrogen reduction figures are only comparable when the crop, the baseline year and the measurement basis all match. They usually do not.
What farmers actually changed
We asked. The answers were consistent enough to be worth listing, and they are more interesting than the headline number.
Accounting for what was already there. Soil mineral nitrogen testing and pre-crop nitrogen credit moved from occasional to routine. A significant share of the reduction is simply not applying nitrogen that was already in the soil.
Rotation. Legumes ahead of wheat — peas in particular — reduce the wheat nitrogen requirement directly. This is old agronomy that became economically attractive again once nitrogen was constrained.
Catch crops. Adopted to reduce over-winter nitrogen loss, though uptake remained limited by cost and the short window between harvest and drilling.
Soil organic matter. Attention shifted towards water-holding capacity and soil structure, particularly after the 2018 and 2019 drought years.
Spatial variability. Growing recognition that yield correlates more strongly with soil than with nitrogen input, which makes uniform application obviously wasteful and precision application obviously sensible.
Cover crops, legume rotation, soil organic matter, reduced input intensity. These are the constituent practices of regenerative agriculture, adopted for economic and regulatory reasons by farmers who were not describing them that way.
“Almost nobody we spoke to said they had adopted regenerative agriculture. Most of them described a set of changes that add up to exactly that.” — Spacenus team
On nitrogen use efficiency
Nitrogen use efficiency improved materially over this period, but the figure depends entirely on how it is defined and this is where a lot of published comparison falls apart.
Measured as nitrogen removed in harvested grain against nitrogen applied, efficiency on these farms rose from well under 70% to above 85%. That definition excludes soil nitrogen supply and mineralisation, so it flatters the result, it is a useful trend indicator, not a mass balance.
Farm-gate nitrogen balance tells a different and equally valid story: a surplus of roughly 80 kg/ha against something closer to 120 kg/ha two decades earlier. Lower, still substantial, and a reminder that there is further to go.
Both are true. Quoting one without naming the definition is how nitrogen debates become unproductive.
What this suggests about regulation
The reflex in agriculture is that environmental regulation costs yield. In this case it did not, and the reason is worth understanding: the regulation targeted an input that was genuinely being over-applied relative to crop requirement, at a point when the tools to apply it more precisely had become available and affordable.
That combination is not universal. The farmers we worked with were clear that blanket regional rules are the weak part, a uniform restriction applied across a region ignores that fields within it differ enormously in what they actually need. Their objection was never to the principle of limiting nitrogen. It was to being told the same number as a neighbour with different soil.
Which is an argument for better measurement rather than looser rules. If you can show which fields are over-applied and which are not, the regulation can be specific rather than blunt, and it becomes a great deal easier to defend to the people it applies to.
ANA delivers the satellite-calibrated nitrogen recommendations behind this research, and continues to run on German farms. There is more on the farmer-facing side of this work at ana.farm.
Common questions
Did the Düngeverordnung reduce yields?
Not on the farms in this study. Winter wheat yields held broadly steady while application rates fell substantially, which suggests the reduction came out of surplus rather than out of crop requirement.
How much of the 2022 reduction was price rather than regulation?
This data cannot separate them. Fertiliser prices rose steeply in 2022 at the same time as regulatory pressure continued, and both push in the same direction. Anyone attributing the change confidently to one cause is going beyond the evidence.
Is 85% nitrogen use efficiency realistic?
It depends on the definition. Measured as grain nitrogen removal against nitrogen applied, and excluding soil nitrogen supply, yes. On a full farm-gate balance the picture is less flattering, with a surplus of roughly 80 kg/ha remaining.
Does this apply outside Germany?
The mechanism is general — where an input is being over-applied and better measurement becomes available, reduction is possible without yield loss. The magnitude is specific to German soils, crops and starting practice.