On-Farm Experimentation: a global movement to accelerate the transformation of agriculture

Experimenting on farm has been part of the research toolkit for nearly two centuries and is now being reinvented as part of a growing shift towards open innovation practices. An international team of scientists has come together to explain how On-Farm Experimentation, or “OFE”, has attracted participants around the world and why the approach differs from previous research practices, arguing that it has the capacity to transform agriculture.

Over 30,000 farms are involved in OFE: what are the principles behind it?

OFE embeds experimental research within the real-world activities of individual farms through mutually beneficial collaborations between farmers, researchers and other members of the agri-food sector. In practice, OFE initiatives take the form of a step-by-step process in which the farmer and scientist define together an investigation’s terms of reference and set up experiments that are tailored to the particular circumstances of the farm.

The team reported on experiences gained from 11 OFE initiatives around the world and estimated that over 30,000 farms are involved in OFE, even though the approach has not yet been formalised and has only rarely received institutional support in a systematic way.

Despite the sheer diversity of goals, approaches, sociotechnical ecosystems and geographical locations involved, the team was able to identify six guiding principles for OFE:

• Real systems – experiments are conducted on farm and are embedded in farm management;
• Farmer-centric – experiments are driven by the farmer’s questions and are performed collaboratively involving, as a minimum, both farmer and scientists;
• Evidence-driven – experiments are based on the analysis of farm-specific data, which may be facilitated (though not dictated) by digital technologies;
• Specialist-enabled – experiments draw on the contributions of external experts, making it possible for new tools to be introduced and varied viewpoints to be considered;
• Co-learning – experiments are built around ongoing discussion between the participants who, in designing and carrying out the experiments jointly, share their visions and experiences, learning from each other and further developing ideas together;
• Scalable – experiments create knowledge that is valuable locally for individuals, and that is also intended to stimulate broader insights.

A pragmatic mechanism to foster locally-relevant knowledge and renew research questions

As we engage in transitions for our agricultural systems through agroecology, data technology, food, One Health etc., it is important to establish a better understanding of the transformational potential of OFE, whose principles incorporate those of open innovation.

Many approaches, past and present, have used frameworks similar to that of OFE, notably farmer-participatory research. What distinguishes OFE is the value it creates for each participant: farmers can explore practices directly related to their own farming needs, while researchers gain access to field data and informal knowledge that enrich their research and open up new pathways. Different knowledge types – produced not only by scientists and farmers but also by advisors and other industry partners – are jointly put to work, making OFE a pragmatic mechanism to accelerate the creation of locally-relevant and directly applicable knowledge.

In France, OFE can be seen in operation in the DEPHY farm network, with nearly 3000 farmers working to test out pathways for the reduction of pesticide use in partnership with expert advisors and researchers. Another project, the Occitanum Living Lab, was recently set up in the Occitanie Region. Coordinated by INRAE and involving 48 partners, Occitanum supports farmers in testing digital technologies and helps them to work in partnership with researchers to assess their economic, environmental and social performance.

The six OFE principles bring into focus the profound changes that have occurred in the ways agricultural experimentation is conducted. Not just a tool to validate lab-based theories, experimentation is becoming a genuine method to stimulate the production of local and applied knowledge. The process renews research questions: how do we hybridise plural knowledge (formal, informal, local, generalisable, data-mined)? What is the nature of the data that supports the construction of knowledge? How do we measure the value it creates, and distribute this value fairly? To address these questions and to boost the fundamental transformational changes that agricultural systems need, the team argue that the OFE movement and its community must be supported and strengthened.