The #DigitAg research has been designed as a matrix: on the one hand, the scientific community is organized into 6 axes. On the other hand, we have identified 2 major societal issues related to agriculture, and 8 associated challenges. The latter form an operational basis for responding to these two societal issues.

Marie Gosme (Inrae) leads the challenges of Stake 1, and Danielle Galliano (Inrae) the challenges of Stake 2

Stake 1 - Improving agricultural production through the use of Information and Communication Technologies (ICT)

Challenge 1: ICT and the agroecology challenge

Agroecology is about using biodiversity and ecological processes to design efficient, sustainable and productive agricultural production systems. Managing more complex systems with solutions adapted to local conditions requires collecting and processing a large amount of information on their biological and physical components (plants, animals, soil, climate). Digital technologies thus appear to be an effective, but still underestimated, lever for the agroecological transition. Meeting this challenge requires the development of: new sensors (axis 3), new information systems and new data processing techniques to connect local and global data (axis 4 and axis 5), new models to predict and support decision-making (axis 5 and axis 6). The answer to this challenge also relies on cooperation with social sciences (axis 1 and axis 2) to connect processes, techniques and people, for example through a living laboratory project.

An example of research would be to study, based on case studies, how ICT can be used to continuously monitor and evaluate ecosystem services in order to help farmers achieve their agroecological transition, while maintaining their cropping system balanced between production level and ecosystem services.

Systèmes Méditerranéens (© UMR AbSyS)

Challenge 2: Digital solutions to optimize the genotype in changing production systems and markets

High-throughput phenotyping of plants and animals is developing rapidly and is an effective tool for implementing new production systems adapted to local conditions. However, until now, sensors, data storage, data processing solutions and models for phenotyping have been designed independently, which makes the phenotyping chain too inefficient and limits its use outside the laboratory.

Tools and methods need to be improved and standardized. For example, a better understanding of physiologies must be introduced in the development of new sensors (axis 3) and models (axis 6). In parallel, an integrative approach is to be used for data and knowledge mining and reuse (axis 2, axis 4 and axis 5). This digital integration of phenotyping tools and methods, coupled with an understanding of farmers' and markets' needs (axis 1) will allow to modernize operational research for the adaptation of ideotypes, in silico, to new production systems and markets.

Research examples:

• How genotype-to-phenotype models, calibrated by employing high-throughput phenotyping information, can be used to inform varietal choices or to develop varieties more adapted to future cropping systems;
• How ICT can help link data collected in farmer networks with genotypic information, to optimize varietal choices and crop management.

Challenge 3: ICT and crop protection

Réduire l'utilisation des pesticides est un défi majeur qui relève à la fois de la transparence, de la sécurité sanitaire, de la préservation de l'environnement, et de la sécurisation du revenu de l'agriculteur. Des barrières techniques, sociologiques, économiques et organisationnelles empêchent des changements majeurs au niveau des pratiques de protection des cultures et la réduction de l'utilisation de pesticides.

De nouvelles perspectives sont requises. Nous nous proposons de travailler avec une nouvelle hypothèse basée sur la co-construction scientifique interdisciplinaire. Elle est basée sur le fait que "avec la protection des cultures, un agriculteur cherche un revenu et un marché", et donc que "la protection des cultures doit être repensée à travers des services adaptés fondés sur les TIC qui combinent des assurances techniques et économiques". La conception, l'organisation et la mise en oeuvre de tels services se base sur une compréhension de l'aversion aux risques individuels et collectifs, la mise en place de services personnalisés (axe 2), le développement de capteurs adaptés aux besoins (axe 3), des dispositifs de traçabilité (axe 1 et axe 4), des systèmes d'information (axe 4) et des modèles basés sur les données et l'expertise (axe 5 et axe 6).

Un premier objectif de recherche serait d'étudier, concevoir et expérimenter de nouvelles technologies d'évaluation du

Reducing the use of pesticides is a major challenge that has to do with transparency, health safety, environmental preservation, and securing the farmer's income. Technical, sociological, economic and organizational barriers prevent major changes in crop protection practices and the reduction of pesticide use.

New perspectives are required. We propose to work with a new hypothesis based on interdisciplinary scientific co-construction. It is based on the fact that "with crop protection, a farmer is looking for an income and a market", and therefore that "crop protection must be rethought through adapted ICT-based services that combine technical and economic assurances". The design, organization and implementation of such services is based on an understanding of individual and collective risk aversion, the implementation of personalized services (axis 2), the development of sensors adapted to needs (axis 3), traceability devices (axis 1 and axis 4), information systems (axis 4) and models based on data and expertise (axis 5 and axis 6).

A first research objective would be to study, design and experiment new risk assessment technologies based on the combination of information from people and sensors, at the field and regional bio-climate scale.

Challenge 4: ICT and sustainable animal production

Animal production is one of the most advanced agricultural sectors in terms of ICT, with new livestock technologies (RFID chips, sensors and milking or feeding robots). It could also benefit from technological advances in connected health. This influx of technology is not only improving the productive efficiency of livestock, but also changing the profession of animal husbandry by creating new links with the animal, and inducing the need for a massive transition of the livestock sector to integrate these new approaches.

While breeders use these devices for monitoring, the various actors in the field of advice, selection, processing, and health institutes could benefit from this information for broader purposes. This raises questions concerning: data quality, interoperability (axis 4), the evolution of current models to better integrate individual and longitudinal dimensions of data in livestock management (axis 5 and axis 6), new multi-criteria objectives for livestock performance (axis 1 and axis 2). Finally, in terms of social and economic issues related to the quality of life of livestock farmers, the economic viability, attractiveness and acceptability of these tools must be studied according to the type of livestock farming system promoted (industrial vs. agro-ecological) (axis 2).

A first example of research would be to develop a decision support tool to manage precision feeding of livestock and improve efficiency in livestock farming.

Stake 2 – A better society inclusiveness for ICT-enabled agriculture
 

Challenge 5: ICT and new farm advisory services

The agricultural advisory sector will be profoundly altered by the diffusion of advisory systems employing digital technologies...

First, innovations are expected, with more personalized and targeted advice based on massive data flows analyzed in real time.

Secondly, ICT will have an impact on the organization of advisory services, which will evolve either towards local networks of actors in collaboration with simple ICT devices, or towards large advisory firms that manage substantial investments to collect and process data.

Finally, the problems related to intellectual property and the consequent sharing of value will be raised.

ICT can lead to great discrepancies between farmers, some of whom are stimulated by the use of these new technologies, while others risk becoming dependent on large companies to collect and process data.

To understand and anticipate these problems, it will be necessary to call upon the scientific communities, for example those of law, management and social sciences (axis 1 and axis 2), computer and information sciences (axis 4 and axis 5) and agronomy (axis 6).

An example of research would be the analysis of new information and communication technologies providing advice at the farm level, and the changes in practices and decision making, according to the specificities of the farm.

Challenge 6: ICT and agricultural territory management

Data produced by farmers and other actors acting on the same territory can help us optimize the use of collective resources for agricultural interests. For example, data acquired through a crowdsourcing campaign can be used to support agricultural actions and/or define systems for monitoring the evolution of the territory.

How to improve data collection (axis 2 and axis 3). How can we organize, manage (axis 4), exploit and share (axis 5) the knowledge hidden in these data? These major challenges must be addressed in order to support the development of territorial information in which the farmer plays an essential role from which he can benefit.

An example of research would be the study of data collected by farmers about their land and their practices (technical itineraries, presence of species of ecological interest, ...) so that land managers can assess the ecosystem services provided by the cultivated areas. Another example would be the study of digital facilitation of the social integration of farmers in the governance structures of the territory.

Challenge 7: Integrating agriculture into value chains

At the production level, advances in automation, robotics, drone and remote control technologies are increasingly enabling the use of remote-controlled robotic solutions or the automation of certain practices; a situation that is changing jobs and altering the "traditional" vision of the farmer and agriculture within value chains.

In terms of sales, e-commerce and m-commerce (digital applications for cell phones) offer retail and catering services with home delivery; a new context that modifies the concept of "market" as a physical place where individuals enter into contact to exchange, negotiate and carry out a transaction, but also transforms the way of communicating with buyers and consumers of agricultural products.

With regard to traceability and transparency on prices, ecological footprint and health quality of products, blockchain technologies offer secure access to increasingly large and complex information; an opportunity that also raises many challenges such as, for example, the use and valuation of this information while controlling its dissemination and limiting unwarranted appropriation by third parties.

These different societal challenges require a multidisciplinary approach that takes advantage of the skills needed to improve information systems (Axis 4) and to extract knowledge from large amounts of data (Axis 5), but also of the approaches and analytical frameworks that the human and social sciences can offer (Axes 1 and 2) to enable a better positioning of farmers and agriculture in value chains and to respond to the expectations of the different actors in the sector (producers, consumers, agri-food industries, retailers, etc.).

One example of research would be the development of digital media capable of connecting producers and consumers while reducing the environmental impact of direct sales or short circuits of agricultural products.

Another example would be the design of management tools that are not limited to inventory tracking and farm accounting, but that also explore available data and use the power of analytics to provide technical, economic, financial and strategic decision support solutions to farmers.

Challenge 8: ICT and agricultural development in Southern countries (Africa)

Southern countries are very interested in digital agriculture, and are constrained by different constraints than Europe. There are many experiences with agricultural advisory services, insurance, credit, and education, but much remains to be done at the innovation level. In particular, a holistic approach could lead to a better understanding of the key factors leading to the success of digital agriculture in the South. This requires research to facilitate the development of services based on "frugal" ICTs such as mobiles and smartphones, internet (and audio blogs), and low cost satellite imagery.

Specialists in innovation (Axis 2), data acquisition technologies (Axis 3), information systems (Axis 4), and modeling (Axis 6) must work together to this end, in order to define a model of ICT-enabled development that meets the needs of both literate and illiterate farmers.

An example of research would be the analysis of new services to farmers provided by advisory organizations, and the design of an economic model that supports such services.