Post-doctorate positions 2021


In 2021, #DigitAg supports post-doctorate positions, based on interdisciplinary topics.

CAUTION: the applicant must not be:

  • a former PhD cofunded or labelled by #DigitAg
  • a former PhD framed by a #DigitAg Associate researcher

The following topics have been selected by #DigitAg.
For further information, please send a mail to the mentionned contact in the post-doctorate offer you are interested in.

Topics available in 2021 :

Life & Environmental Sciences + Applied Mathematics 

Added value of farmers’ weather IoT stations and ensemble weather forecasts to provide  local advice on water management integrating uncertainty

Keywords: agro-meteorology, decision support tool (DST), internet of Things (IoT), uncertainty, water balance

  • Contact: François Brun – francois.brun[AT]
  • Supervisors: François Brun, Acta -Sébastien Roux, Mistea, Inrae – Laure Raynaud, CNRM, Météo-France
  • Research units: Acta, Mistea
  • Co-funding : Acta
  • #DigitAg: Axis 6, Axis 3, Challenge 0
    Challenge 1,Challenge 5

Agriculture is one of the activities most impacted by meteorological hazards, modulating the crop cycle, irrigation management, crop protection and fertilization. This sector has a strong demand for decision support tools (DST) aimed at adjusting crop operations according to weather conditions. To this end, more and more farmers are acquiring and installing IoT weather stations in order to supply the DSTs with in situ meteorological observations as close as possible to their farms. Nevertheless, the deployment of these new low cost stations raises the question of their quality, their implantation, their maintenance and the procedures to complete the missing data. At the same time, in recent years, weather forecasts have evolved, now offering information on uncertainty. These probabilistic forecasts, also called ensemble forecasts, allow to propose realistic scenarios to represent the uncertainties of forecasts up to 15 days. All these elements participate in the construction of a precision agriculture.

The objective of this post doc is to carry out work on this subject, along two Axiss. The first axis will aim to highlight the contribution of algorithms for processing the measurements of the IoT stations set up in the METEOPREC project. In addition to their usefulness for maintenance, these algorithms will make it possible to secure the use of these measurements, particularly in DST for farmers. Indeed, the latter are sensitive to errors on key variables such as temperature, relative humidity or rainfall. The second axis will describe the joint use of station weather data with ensemble forecasts applied to two DST for irrigation management in vineyards and corn. It will show the interest of ensemble forecasting approaches compared to classical forecasts or frequency approaches. It is planned to write two scientific articles to cover this work.

The use of on-board sensors to model livestock behaviour and pathogen transmission : An application for PPR in Senegal

Keywords: onboard sensors, animal movement, multiscale modeling, epidemiology, companion modeling, Subsaharan Africa, livestock husbandry.

  • Contact: Jean Baptiste Menassol – jean-baptiste.menassol[AT]
  • Supervisors: Jean Baptiste Menassol, Selmet, Institut Agro – Montpellier Sup Agro -Andrea Apolloni, Astre, Cirad – Maxime Lenormand, Tetis, Inrae
  • Research units: Selmet, Tetis, Sens
  • Co-funding: To be found
  • #DigitAg: Axis 6, Axis 1, Axis 3
    Challenge 4, Challenge 1,Challenge 6,Challenge 8

The use of telemetry and on-board sensors is expanding rapidly in the field of livestock science and opens up new perspectives for modeling and preventing systemic risks such as epidemics. Contacts between animals play a fundamental role in the transmission and spread of animal diseases such as Peste des Petits Ruminants (PPR). In rural areas of sub-Saharan Africa, animal interactions can be analysed at three scales:

(1) between individuals of the same herd;

(2) between herds sharing the same pastoral resources(rangelands, waterpoints, markets, etc.);

(3) occasionally with transhumant animals, including cross-border movements.

The work of the postdoc will integrate datasets from ongoing research projects in Senegal (ecoPPR, DSCATT, CASSECS). Data from sensors inform the different scales of interaction: (i) RF sensors for contacts within a herd during an experimental infection, (ii) accelerometers and GPS to infer behavior during daily trips (iii) herd management rules described by herders during participatory modeling workshops. All of these data will be combined in a model that will reproduce livestock movement patterns in pastoral areas, their interactions and simulate the fine mechanisms of PPR virus transmission within this population. Different scenarios involving variation in pastoral resources and mobility management will be explored to reduce the risks of transmission while preserving the integrity of pastoralists’ mobility practices

Humanities & Social Sciences + Life & Environmental Sciences

Determinants, modalities and value of data sharing by farmers and SMEs in blockchains for transparent and sustainable food supply chains

Keywords: Blockchain, traceability, transparency, food supply chain, power, data sharing, digitalisation, valorisation, participatory sciences

  • Contact: Florent Saucède – florent.saucede[AT]
  • Supervisors: Florent Saucède, Moisa, Institut Agro – Montpellier SupAgro -Léa Tardieu, Tetis, Inrae
  • Research units: Moisa, Tetis
  • Co-funding: Project ANR JCJC
  • #DigitAg: Axis 1, Axis 2,Axis 4
    Challenge 7, Challenge 5

Consumers’ mistrust of complex food systems increases with the number of food scandals and sanitary crises. To address this, producers and retailers are experimenting with blockchain, “a digital, decentralised and distributed ledger in which transactions are recorded and added in chronological order with the goal of creating permanent and tamperproof records” (Treiblmaier, 2018, p. 547). It offers a novel way to trake and trace products along food supply chains (FSCs), thank’s to the members’ contribution and joint construction of immutable information that can be visible to all and communicated to consumers. It has the potential to improve the functioning, digitisation, automation and sustainability of FSCs. While blockchain allows for the design of new collective and participative modes of organisation, it is also a monitoring system whose transparency is co-constructed through the sharing of sensitive data previously kept private. The conditions to enable these potentials and minimise the risks of this technology are therefore unknown.

The post-doctorate is part of an ANR JCJC project proposal that assesses the potential of blockchain to make FSCs more participative, transparent and efficient, to contribute to the transition towards sustainable food systems. Focusing on farmers and SMEs, the post-doctorate aims to better understand the conditions, modalities, risks, reluctances, and individual and collective costs and benefits of data sharing in FSCs for traceability, transparency and valorisation of practices. The determinants and valorisation of data sharing are examined in the context of power dynamics within FSCs. Mobilising management, economic, environmental and data sciences in a participatory approach with farmers and SMEs, the project aims to develop a grid structuring the co-construction of information for sustainable, transparent and efficient FSCs, and to identify the data necessary for its construction, while helping to prepare these producers for the challenges of the spread of such systems of FSCs’ transparency.

Life & Environnemental Sciences + Engineering Science 

Deep4mix : using deep learning to monitor vegetation species dynamics in mixed crops

Keywords: Agroecology, instance segmentation, semantic segmentation, automatic vegetation characterization

  • Contact: Marie Weiss – marie.weiss[AT]
  • Supervisors: Marie Weiss, Emmah, Inrae – Alexis Joly, Lirmm, Inria- Lionel Alletto, Agir, Inrae
  • Research units: Emmah, Lirmm, Agir
  • Co-funding: 100% #DigitAg funding
  • #DigitAg: Axis 5, Axis 2
    Challenge 1, Challenge 2

The agroecological transition requires the development and the assessment of new multiperformant, resilient and sustainable agroecosystems. However, these systems are currently lacking high-throughput, non-destructive and objective observation tools. The data deficiency is even more critical because of the higher complexity of agroecosystems such as mixed crops as compared to single crop systems. High-throughput observation tools based on close range imagery therefore appear as essential for rapidly characterizing and hence better understanding these new agrosystems. However, if these tools have now reached a certain maturity for the monitoring of monospecific crops, their use in agroecology remains limited. This project aims to understand to what extent close-range imagery can be used for field monitoring of the dynamics of the proportion and structure of species in a crop mixture.

The methodological approach involves a preliminary step of species identification within the canopy using deep learning models. A new database of annotated images will be therefore first created, and based on data acquired on multispecies crop canopies in the Remix project, but also on data acquired on single crop systems  including weeds or not. The contribution of ancillary information (RGB images in two viewing directions, 3D LiDAR point cloud) as input to the deep models will be also investigated. Finally, once the species are identified, the project aims to estimate new traits such as the proportion of species, the corresponding leaf area or the overlapping area between these species, as well as to use the dynamics of these traits to identify key events such as the date of species cover.

Co-design of participatory simulations of MAS models for multi-actor decision support

Keywords: companion modeling ; multi-actor co-design ; sustainable management of pastoral territories; co-constructed interfaces

  • Contact: Jacques Lasseur – jacques.lasseur[AT]
  • Supervisors: Jacques Lasseur, Sens, Cirad-Alexandre Ickowicz, Selmet, Cirad- Jacques Lasseur, Selmet, Inrae
  • Research units: Selmet, Sens
  • Co-funding: FAO-GASL, MEAE, Projets UE DESIRA and UE LEAP
  • #DigitAg: Axis 6, Axis 2
    Challenge 4, Challenge 8

Simulation models are increasingly used to help better understand the functioning of territories, to project their probable futures, to explore multiple sustainable management options and their possible impacts, and this according to a variety of perspectives carried by a variety of actors. Currently, most models are black boxes with parameters as inputs that describe both the functioning of the territory and the possible options, and time series of indicators as outputs, possibly spatialized, that allow visualizing the impact over time of the options taken. Moreover, the models are often limited to a small part of the functioning of the territory (plot, farm management, supply chain/sector) because integration at the scale of the territory implies hundreds of parameters and indicators, which brings a complexity that is difficult to manage. The use of simulation models is therefore confronted with the requirement of transparency so that the non-scientific actors understand the hypotheses made in the models, their functioning and the interactions between the different processes that take place in them, and with the requirement of usability in a multi-actor context, i.e. that each actor is provided with a view adapted to his objectives in terms of the options and indicators at his disposal. The objective of this project is therefore to design the methods and the technical means to surround existing models at the scale of the territory with computer tools and graphical interfaces allowing a variety of actors to understand their functioning and to have the means to explore a variety of scenarios (possible futures) and options (possible decisions) and to evaluate their impacts. This implies upstream the identification of the different actors concerned, their stakes and objectives and the knowledge they need, and downstream the implementation of tools for the manipulation and visualization of models adapted to their needs. The methods and tools developed will be involved in the governance platforms of territories in the North as well as in the South and will serve as a support for discussions between stakeholders on the future of their territories.

LADi D: Long-term Analysis of Dieback Drivers.   Analysis of the determinants of multi-year vineyard dieback trajectories  by Bayesian linear regression for functional data

Keywords: Bayesian linear regression, functional data, vineyard dieback, agricultural transition trajectories.

  • Contact: Meili Baragatti – meili.baragatti[AT]
  • Supervisors: Baragatti Meili, Mistea, Institut Agro -Hilgert Nadine, Mistea, Inrae- Nathalie Smits, ABSys, Inrae
  • Research units: Mistea, ABSys
  • Co-funding : LabEx NUMEV
  • #DigitAg: Axis 5
    Challenge 1, Challenge 3

In recent years, the wine industry has been complaining about a drop in yields and vine longevity. Vineyard dieback is a syndrome that results in a drop in productivity over several years and the premature death of certain vines. A particular feature of vine dieback is that it occurs over long periods of time, from a few years to a few decades. Few data are available over such long periods.

The Bureau National Interprofessionnel du Cognac (BNIC) has monitored 55 plots since 1977. This unique database offers the opportunity to analyse the determinants of multi-year vine dieback trajectories, notably by undertaking a “dynamic” characterisation of vine yield and mortality (i.e. with explanatory variables that evolve over time).

The objective of this post-doc project is to identify (1) the factors and interactions of biotic, abiotic and technical factors that contribute to the decline in plot yield and to the mortality of individual grapevines and (2) the time period over which these factors have an impact, both in the short term at the scale of the crop cycle and in the long term since the plot was planted. Exploratory statistical analyses will be required, as well as methodological developments in linear regression for functional data.

Beyond the analysis of this case study, the post-doc will produce a generic protocol for the analysis of temporal data on agricultural systems that will fill a gap in statistical tools for the analysis of agroecological transition trajectories.

Applied Maths + Engineering Science

Transitool: a companion tool to assess step-by-step agroecological transition of a cash crop farm

Keywords: agroecological transition, simulation, multicriteria assessment, arable crops, ergonomy, AGILE approach

  • Contact: Jacques-Eric Bergez – jacques-eric.bergez[AT]
  • Supervisors: Jacques-Eric Bergez, Agir, Inrae – Isabelle Bourdon, MRM, Université de Montpellier – Patrick Taillandier, Miat, Inrae
  • Research units: Agir, MRM, Miat
  • Co-funding: 100% #DigitAg funding
  • #DigitAg: Axis 6, Axis 2
    Challenge 1, Challenge 5

Switching from conventional practices to more agroecological practices is a real challenge for the farmers. This transition can be done step-by-step or using a complete system redesign. Whatever path is chosen, fostering initial reflection between advisers and farmers is an approach that has the big interest of creating a community of exchanges and innovative practices. The Transitool project is interested in the implementation of a companion tool that can be mobilized in a co-design process of more agroecological technical practices for arable farms. This is to help the various actors of a co-design workshop to assess changes in practices through the use of computer simulation to quantify various indicators showing changes in the farm. The project aims to meet three major challenges: i) develop a useful and usable tool, ii) develop a tool allowing a multi-criteria assessment (sustainability), iii) develop a tool mobilizing recent agroecosystem simulation approaches. To meet the first challenge, the post-doctoral fellow will follow an AGILE approach to develop, via use-cases, interfaces facilitating the work of co-design and evaluation of the practices tested. To meet the second challenge, the post-doctoral fellow will mobilize multi-criteria assessment procedures for cropping systems (eg MEANS, MASC). Finally, to meet the third challenge, the post-doc will use tools and programming language already mobilized by researchers (eg MAELIA, GAMA) and will seek to hybridize quantitative and qualitative simulation approaches. In order to properly mobilize expectations and skills, a steering committee including researchers in the field, but also professionals from the agricultural world will be set up.