7 Partners
4 Agronomists, 2 Universities, 1 Tech partner and 60 farmers.
PRACTICE, SCIENCE AND TECHNOLOGY
Different expertise. One agronomic challenge.
Understanding soil and crop health requires more than one discipline. The project brings together agronomic practice, regenerative agriculture, soil biology, digital technology and artificial intelligence.
Data, platform & agronomic technology
SoilBeat
SoilBeat provides the digital foundation of the project. The platform connects soil and crop information, integrates research outcomes and helps turn complex measurements into structured agronomic context.
Platform · data integration · agronomic intelligence
Visit website
Soil biology & microbial ecology
University of Groningen
The University of Groningen investigates whether enzymatic activity of soil microorganisms can contribute to a better understanding of soil functioning and soil quality.
Soil research · bio-indicators · scientific validation
Artificial intelligence & machine learning
Vrije Universiteit Amsterdam
Vrije Universiteit Amsterdam contributes expertise in artificial intelligence, machine learning and data analysis to investigate patterns across combined soil and crop datasets.
AI · machine learning · model development
Regenerative agriculture & farmer guidance
CO2L Farming
CO2L Farming connects the research and technology in the project with regenerative agricultural practice and supports farmer participation, learning and implementation.
Regenerative farming · training · farmer guidance
Agronomy & practical interpretation
GroeiBalans
GroeiBalans brings practical agronomic expertise in soil and crop management and helps participating farmers interpret measurements within their own crop and field context.
Agronomy · plant sap interpretation · farmer guidance
Soil health & practical agronomy
Mulder Agro
Mulder Agro contributes practical experience in soil health, regenerative principles and farm advisory work to the interpretation and validation of measurements.
Soil health · agronomic advice · practical validation
Crop resilience & field agronomy
WeerBaar Telen
WeerBaar Telen contributes practical knowledge of resilient crop production and helps connect measurements with crop development and field observations.
Crop resilience · plant sap interpretation · agronomy
FROM RESEARCH TO PRACTICE
What the project is building, testing and learning.
The project is deliberately broader than software development. New measurements only become useful when the science is robust, the data can be interpreted and farmers can apply the resulting knowledge in practice.
01
Platform
A further developed SoilBeat platform
New functionality connects soil and crop measurements with agronomic knowledge, including contextual microlearnings, structured interpretation and relevant data integrations.
02
AI & Data
AI and machine-learning models
Combined datasets from soil measurements, plant sap analyses and biological indicators are prepared, analysed and used to develop and validate machine-learning models.
03
Research
Soil bio-indicator research
The University of Groningen investigates enzyme activity of soil microorganisms and its potential value as an indicator of soil functioning and soil quality.
04
Field validation
A practical pilot with participating farmers
Farmers use the measurements, SoilBeat and regenerative principles in real production environments, creating feedback and practical validation across multiple growing seasons.
05
Agronomy
Measurements plus agronomic guidance
Plant sap, soil and biological measurements are combined with agronomic guidance so that individual values can be interpreted within crop, field and management context.
06
Learning
Continuous learning throughout the project
Insights from research and field experience are fed back into the platform and its knowledge layer rather than waiting until the end of the project.
07
Knowledge sharing
Sharing what the project learns
Project knowledge is shared with farmers, advisors, education, agricultural organisations and wider audiences through demonstrations, events, publications and other knowledge-sharing activities.
FROM RESEARCH QUESTION TO FIELD VALIDATION
A project built through repeated cycles of measuring, learning and improving.
Platform development, scientific research and practical field validation run in parallel. That allows new evidence from farms and research to flow back into SoilBeat throughout the project rather than only at the end.
2023
Project preparation and development start
The consortium establishes the research, technical and practical foundations of the project, including platform development, measurement protocols and preparation for field participation.
Project foundation
2024
Field validation begins
Participating farms begin using the measurement methods and SoilBeat in practice. Plant sap, soil and biological measurements start creating a shared evidence base across the project.
First field cycle
2024–2025
Research and data models develop
Soil research, practical measurements and combined datasets are analysed while AI and machine-learning models are developed and tested against the growing evidence base.
Research + model development
2025
Repeated measurements add context
Additional measurement cycles make it possible to compare patterns through time rather than treating every laboratory result as an isolated snapshot.
Longitudinal learning
2026
From project learning to wider application
Project knowledge, agronomic experience, research outcomes and technology are brought together so that the lessons can be translated into practical tools, interpretation and wider knowledge sharing.
Knowledge into practice
FOR FARMERS
Turn your measurements into better decisions on the farm.
Every field tells a story, but it is often spread across soil analyses, plant sap results, crop observations, weather, applications and years of experience. By participating in the SoilBeat SNN project, you help us bring those pieces together — while gaining a clearer view of what is happening in your own soil and crop.
01
See what is happening inside the crop
Plant sap analysis provides a current view of nutrient uptake and plant status. Repeated measurements help you follow how the crop responds during the season instead of relying on one isolated result.
02
Look beyond a single measurement
Soil measurements, plant sap results, biological indicators and field observations are brought together in SoilBeat so that relationships and trends can be viewed in context.
03
Discuss the results with agronomists
Measurements are most useful when they lead to better questions. Participating farmers receive support in interpreting results and identifying what should be checked before making changes in the field.
04
Learn across multiple seasons
Repeated measurements make it possible to compare crops, seasons and management decisions. Each new measurement adds context to the next one.
What will we ask from you?
Participation is designed to fit around normal farm management. You do not need to become a data scientist — the purpose of the project is to make complex agronomic information more practical to use.
- Take or provide soil and plant samples at agreed moments.
- Record relevant crop, fertilisation and field activities.
- Discuss results with participating agronomists.
- Use SoilBeat to keep measurements and farm context connected.
- Share practical feedback on what is — and is not — useful on the farm.
Why your participation matters
Research can show what is scientifically possible. Farmers help determine what is practically useful. By participating, you help us test whether new measurements, agronomic interpretation and digital tools actually improve decision-making under real farm conditions.
Which measurements provide useful early signals?
Which patterns are repeatable across seasons?
What information helps before an input decision is made?
How should complex laboratory results be presented so they are useful in practice?
Interested in participating?
Tell us a little about your farm, crops and location. We can then discuss whether your farm fits the current project activities and what participation would involve.
FROM MEASUREMENT TO AGRONOMIC INTERPRETATION
A plant sap value tells you what was measured. SoilBeat helps you understand what it means.
A plant sap analysis contains valuable information, but a single value rarely tells the whole agronomic story. A low value in young leaves can mean something different from the same value in old leaves. A change over time may look important, but only if the samples are genuinely comparable. And an abnormal nutrient value may be related to uptake, root activity, crop development, water availability, salinity, or interactions with other nutrients.
That is why SoilBeat interprets a plant sap analysis as an agronomic pattern — not as a collection of isolated numbers.
THE INTERPRETATION LOGIC
From laboratory result to agronomic interpretation.
SoilBeat builds the interpretation in layers. Each step adds context before a conclusion is drawn.
SoilBeat does not begin with the relationship between the leaves. Each measured value is first assessed independently against the relevant laboratory or crop reference range. This keeps the young-leaf status, old-leaf status and the direction of both signals clearly visible before their relationship is interpreted.
A balanced relationship between young and old leaves does not automatically mean nutrient status is sufficient. Both values could still be low.
Not every nutrient moves through the plant in the same way.
The mobility of an element influences how much weight young leaves, old leaves and their distribution receive in the interpretation. SoilBeat therefore uses different interpretation groups.
Mobile
N · NO₃-N · NH₄-N · P · K · Mg
Young and old leaves both provide useful information. Their distribution can add context around redistribution, reserve depletion or relative accumulation in older tissue.
Limited mobility
S · Fe · Mn · Zn · Cu · Mo
Young leaves generally carry more weight when assessing current status, while old leaves remain important physiological context.
Young leaf primary
Ca · B
The current status in young tissue is decisive. A high value in old leaves does not automatically compensate for a low value in young leaves.
Specific interpretation
Cl · Na · Si · Al
These elements require more element-, crop-, laboratory- or context-specific interpretation rather than a generic mobility rule.
Where a nutrient is found can be as informative as how much is present.
The distribution between young and old leaves can add information about redistribution, reserve depletion, relative accumulation or unusual patterns. SoilBeat always combines that signal with the absolute values, crop, growth stage and other available context.
Nutrients function as a system, not as independent sliders.
When combinations of measurements suggest a possible interaction, SoilBeat can surface that pattern for further investigation. The signal is treated as a hypothesis and a verification point — not proof of causality.
K ↔ Ca / Mg
High K together with low Ca or Mg may warrant investigation of possible competition between cations and the wider root-zone context.
P ↔ Zn
High P together with low Zn may warrant investigation of a possible P–Zn interaction.
NH₄-N ↔ K / Ca / Mg
High ammonium combined with low cation values may justify checking for uptake competition or other limitations.
Cl ↔ NO₃-N
High chloride combined with low nitrate may prompt a closer look at salinity, irrigation water and possible competition.
NO₃-N ↔ S / Fe / Mo
When nitrate is high, other elements involved in nitrogen metabolism can provide useful additional context.
Na ↔ K / Ca
High sodium together with low K or Ca may be consistent with a wider salinity or ion-balance issue.
From signal to verification point — not directly to cause.
SoilBeat does not automatically conclude that high potassium caused a calcium problem. Instead, the interpretation may indicate that the pattern is consistent with possible competition between K, Ca and Mg and recommend checking root-zone conditions, irrigation, EC, recent applications and other available evidence.
AGRONOMIC GUARDRAILS
Interpretation does not stop at ‘low’ or ‘high’.
It is easy for software to flag a value when it falls outside a reference range. Agronomic interpretation requires more. SoilBeat therefore uses guardrails designed to prevent a single datapoint from being translated too quickly into an intervention.
An abnormal value is not automatically a product recommendation.
A low nutrient status does not necessarily mean that more of that nutrient should immediately be applied. Root activity, water, oxygen, EC, pH, nutrient competition, growth rate and recent applications may need to be checked first.
A ratio never overrides the underlying measurements.
A favourable relationship between young and old leaves can still occur when both values are too low. The individual leaf statuses therefore remain part of the interpretation.
Young tissue remains decisive where physiology requires it.
For nutrients such as calcium and boron, nutrient stored in older tissue cannot automatically be treated as available to developing tissue.
An interaction is a hypothesis, not proven causality.
SoilBeat uses language such as ‘may be consistent with’, ‘could indicate’ and ‘verify’ where the available data cannot prove one specific cause.
Trends require comparable measurements.
A different laboratory, sampling approach, crop stage or leaf position can create an apparent trend that does not reflect a genuine physiological change.
Context comes before correction.
The preferred sequence is to detect, investigate, verify, act when agronomically justified, and then monitor the outcome with new evidence.
Not every interpretation has the same strength of evidence.
SoilBeat can indicate how strongly the available data and context support an interpretation. Confidence relates to the quality of the interpretation — not certainty that one specific root cause has been proven.
Higher confidence
The sample is usable, the pattern is clear, relevant signals support one another and there are few important contradictions.
Moderate confidence
The interpretation is plausible, but some relevant context is missing or credible alternative explanations remain.
Lower confidence
Important uncertainty exists around sampling, comparability, missing context or strongly conflicting physiological signals.
Missing or conflicting context can therefore lead to a more cautious conclusion even when the laboratory measurement itself is valid.
From laboratory report to agronomic reasoning.
SoilBeat helps growers and agronomists ask the same essential questions consistently before deciding what should happen next.
What are we seeing? What do the young and old leaves tell us? How does this nutrient behave within the plant? Does the pattern make sense for this crop and growth stage? Are we looking at a genuine trend? Which interactions or alternative explanations matter? What needs to be verified in the crop, root zone, soil, substrate or water? How strongly does the available evidence support the conclusion?
The Project in Detail
Maximum number of participants: 125 farms
Target group: Cattle farmers, arable farmers and horticulturists focused on soil health and crop growth.
Duration: 2024, 2025, and 2026
Includes: Use of SoilBeat for data organization, personal intake with advisor, multiple plant sap measurements with analysis and advice.