Vous êtes


Réduire la taille texte Rétablir la taille du texte par défaut Agrandir la taille du texte Partager cette page Favoris Email Imprimer


© M. Carrouée

Models, information systems and viable management of the environment

In summary

In a word

  • 16 permanent staff including 15 researchers, engineers and technicians
  • 20 top-ranking publication per year on average
  • patent

Four lines of scientific research

  • Optimising the performance of wireless sensor networks for agri-environmental applications
  • Designing and developing agri-environmental information
  • Modelling ecological and social spatial dynamics
  • Computing action policies to enhance for viability and resilience

Editorial : the challenge of mathematics for sustainable territorial management

Could mathematics and information technology help us improve our understanding and conservation of ecosystems? In fact, the widespread use of networked sensors and the growth of Linked Data are already generating large volumes of easily accessible environmental data (from farms, forests, rivers etc.) That is a real gold mine for researchers. Getting the most out of this data is nevertheless far from straightforward. How are we to take its diversity and spatialised nature into account? How should we manage its uncertainties, especially when it has been collected by non-specialists? How can we make it easier for such quantities of data to be shared and used?

Equally, major advances in modelling methods combined with a continued increase in processing power are opening up new horizons. They effectively allow us to build models from multiple data sources that draw closely on expert knowledge and conditions in the field. The elements of the system are modelled, for example trees in a forest or farmers in a region…, along with rules on how they can change. Our computers then calculate development trajectories for these populations. Such simulations allow us to observe emerging patterns such as the distribution of trees of two different species in order to gain an understanding of the mechanisms involved. Last, new mathematical techniques have made it possible to control and maintain certain properties in a system such as levels of biodiversity for a forest or income for a farmer.

These techniques, derived from viability theory, would seem well-suited to the goals of sustainable development. They also help to clarify the concept of resilience. These exciting scientific advances inform the work of the MOTIVE (Models, information systems and viable management of the environment) research theme scientists in partnership with specialists on methodologies (CNRS, INRA and universities), subject specialists and endusers.

In future the teams will be extending their research on links between data and models. More detailed work in collaboration with LAPSCO (Laboratory of Social and Cognitive Psychology, based in Clermont-Ferrand) on the modelling of social dynamics such as the growth in organic farming will contribute to this.

Selected research projects

Insulation panels made from sunflowers


From sensor to indicator: the EDEN project


Supporting public policy

Agroecology goes wiki!

As part of France’s Ecophyto 2 plan, Irstea and its partners are developing a knowledge-management tool intended to promote greater sustainability in agriculture.

The French government has asked farmers to reduce their use of plant protection products by 50% while ensuring continued economic efficiency. This ambitious target is set out in the Ministry of Agriculture, Agrifood and Forestry’s plan known as Ecophyto 2. But the necessary measures to achieve this (crop rotation, soil improvement, etc.) are often disregarded and are not systematically implemented or shared. A possible solution is provided by GECO, the collaborative tool designed by Irstea in association with ACTA (the network of animal and plant production institutes), Inra and the Burgundy regional chamber of agriculture.

Intended to manage and share technical know-how concerning agroecology in order to promote its use, it will include a ‘wiki’-style webspace with technical factsheets on various agricultural practices and a discussion forum allowing stakeholders (farmers, advisors, trainers, researchers) to share their experiences. A prototype, Agro-PEPS, developed in the course of an Irstea doctoral thesis, has been trialled since 2011. Over 150 factsheets have been created and can be accessed through a multiple-criteria semantic search engine. The final version of GECO will go on line in 2017 and become part of the EcophytoPIC portal, another digital outcome of the Ecophyto 2 plan.

Data capture and management

A super-networked farm
In Montoldre, not far from Irstea’s Clermont-Ferrand centre, a research platform on digital agriculture has been created that is one of a kind.

The challenge of increasing and improving productivity in an environmentally-friendly way has become a physical reality in the field in Montoldre where a farm equipped with sensors makes possible the capture and management of agri-environmental data such as weather conditions, rainfall, temperature and ground moisture levels, weight and diet of animals housed indoors and fuel consumption by farm machinery. This data informs research such as the EDEN project on the optimisation of agricultural equipment to control energy consumption on farms, as well as more basic research on the functioning of sensor networks. The platform designed in partnership with LIMOS (Laboratory of Computer Science, Modelling and System Optimisation) will form part of the future AgroTechnoPôle project at Montoldre offering a range of research, innovation, trials and training on intelligent agricultural equipment and digital agriculture to stakeholders in agriculture.