PhD positions 2020

Every year, #DigitAg launches a PhD campaign and supports new theses, from blue skies to applied research. Discover all our Autumn 2020 PhD opportunities!

PhD positions – Campaign 2020

 

Applied Mathematics – Life & Environmental Sciences

Modelization of maize architecture : estimating radiation interception, using sensors and 3-D models

Keywords: Maize, 3D plant model, Plant-to-plant competition, Radiation interception, Sensors, High throughput phenotyping

  • Contact: frederic.baret [AT] inrae.fr
  • PhD director(s) : Frédéric BARET, EMMAH, INRAE
  • Supervisor(s): Raul Lopez Lozano, EMMAH, INRAE / Benoit De Solan, Arvalis /Jean-Charles Deswarte, Arvalis
  • Research units: EMMAH / Arvalis
  • Co-funding :  #DigitAg – Acta
  • #DigitAg: Axes 6 & 2, Challenges 1, 2 & 5

Optimizing maize crop structure is an important strategy to improve yield. Characterizing the morphological plasticity of maize plants as a response to intra-specific competition is necessary to identify the optimal combination between canopy architecture (crop density and structure) that maximizes crop productivity. In that context, crop architectural and physiological characteristics can be assessed through high-throughput optical observations acquired from plant phenotyping platforms. However, understanding maize morphological plasticity requires the use of 3D plant models, that can integrate architectural parameters derived from optical observations, and can describe canopy light regime. This project is structured around two main goals: 1. To evaluate the consequences of sowing patterns (row width, plant density, orientation) on light interception by the crop and to understand their influence on yield formation. / 2.    To improve the use of sensors (high throughput phenotyping, drones, satellites) in order to obtain a detailed crop architectural description. To achieve both goals this PhD will study the architectural plasticity and the light interception and light use efficiency of a range of maize cultivars, in response to intra-specific competition. These indicators will be related to crop performance. The project addresses 4 objectives: 1. Accessing structural crop parameters through optical data obtained in maize plots at early stages. Several techniques will be used to reconstruct 3D descriptions in order to characterize plant and crop architecture / 2.           Estimating canopy light regime (light quantity and quality) using realistic 3D architectural models and ray-tracing algorithms. / 3. Analyzing and describing plant response (organogenesis, LAI and biomass growth, yield components) using architectural and functional plant modelling. / 4.  Characterizing genotypic plasticity associated to intra-specific competition analyzing the dynamics of canopy light regime and plant architecture.

Applied Mathematics – Life & Environmental Sciences

Metrics to assess the spatialisation of crop models for Precision Agriculture: application to a vine water status model

Keywords: Sensitivity analysis, Geostatistics, Spatial models

  • Contact: james.taylor [AT] inrae.fr
  • PhD director(s) : James Taylor, ITAP INRAE/ Bruno Tisseyre, ITAP, L’Institut Agro – Montpellier SupAgro
  • Supervisor(s): James Taylor, ITAP, INRAE/ Sébastien Roux MISTEA, INRAE/Bruno Tisseyre, ITAP, L’Institut Agro – Montpellier SupAgro
  • Research units: INRAE, ITAP & MISTEA
  • Co-funding :  #DigitAg – INRAE
  • #DigitAg: Axes 6 & 5 , Challenges : cross-cutting subject

There exists enormous knowledge gaps in how to best manipulate crop models to accept and to be updated with spatial (and temporal) high-resolution ancillary data, the implications that changing the model has on predictive power (and subsequent management options), and how to properly assess model performance at varying scales. Agricultural scientists need help achieve this and the crux of this thesis will be method development to generate metrics to assess the effect of incorporating multi-temporal spatial crop and environmental observations into existing crop models. The intent will be to incorporate aspects of spatial variance decomposition into a Sobol-based sensitivity analysis. The intent is to improve understanding of how to spatialize model predictions for enhanced spatial management. It will not and cannot address all issues, but will start to provide tools to achieve this. The intent is not to arrive at the best spatialize model, but to develop tools that will help all models arrive at this point. Spatial crop (agri-environmental) models will generate outputs with a change in extend, coverage and/or support from traditional crop model applications. The need for correct methods of sensitivity analysis has been previously discussed and proposed, but only for large scale, regional applications. High resolution, sub-field, agronomic applications are an area of sensitivity analysis that requires further work.

Engineering Science – Humanities & Social Sciences

Use of agent-based simulation and argumentation framework to better understand the diffusion and appropriation of communicating water meter technology in agriculture

Keywords: agent-based modeling, argumentation theory, innovation diffusion, communicating water meter, agricultural territory

  • Contact: patrick.taillandier [AT] inrae.fr
  • PhD director(s) : Patrick Taillandier, INRAE MIAT
  • Supervisor(s): Rallou Thomopoulos (INRAE IATE) et Stéphane Couture (INRAE MIAT)
  • Research units: INRAE MIAT (Toulouse) / INRAE IATE (Montpellier
  • Co-funding :  #DigitAg – INRAE
  • #DigitAg: Axis 1 : Impact des technologies de l’information et de la communication sur le monde rural, Challenges : cross-cutting subject,

These last years have seen a strong development of digital technologies in agriculture, which have already largely begun to impact farmers’ practices. While these technologies offer a unique opportunity to contribute to the emergence of a more environmentally friendly agriculture, they also raise many questions about the negative effects they could be causing (inequalities between farmers, industrialization of agriculture, data ownership and leakage, etc.). This thesis project aims to study different levers that could allow the appropriation and virtuous diffusion of digital tools in agriculture. To do this, it proposes to build a simulation model to evaluate different policies (training of farmers, communication around digital tools, etc.) at the scale of a territory. The model will be based on the coupling between agent-based modeling and argumentation theory. The objective is to start from the field using methods from experimental economics and surveys, and to analyze the arguments used by the various actors on the use of these tools to propose a rich and realistic modeling of the phenomena of appropriation and diffusion of innovations. This modeling approach will be applied to study the case of communicating water meters (similar to Linky for electricity meters) among farmers in Occitanie. These tools, which could lead to better management of water resources, are now a source of tension for many farmers. The application challenge is to better understand these tensions and evaluate strategies to overcome them.

Engineering Science + Humanities & Social Sciences

Augmented reality to support agroforestry systems design

Key words : augmented reality; embedded systems; agroforestry; system design; visual impact

  • Contact : Marie Gosme  – marie.gosme@inrae.fr
  • PhD Director(s):  Marc Jaeger, AMAP, CIRAD
  • Supervisor(s): Marie Gosme, SYSTEM, INRAE – Marc Jaeger, AMAP, CIRAD – Gérard Subsol, LIRMM, Université Montpellier
  • Research units : System
  • Co-funding: INRAE
  • #DigitAg : Axe 2 : Innovations en agriculture numérique, Axe 6: Modélisation et simulation, Challenge 1 : Le challenge agroécologique, Challenge 8 : Développement agricole au Sud

Agroforestry is recognized as a way of developing sustainable, resilient agriculture and combating climate change. However, the number of species combinations, spatial configurations, tree and crop management options is vast. The choices must be adapted to the pedo-climatic and socio-economic contexts and to the objectives of the farmer, who therefore needs support to design the system. New technologies can facilitate this support and promote the adoption of agroforestry. Augmented reality (AR, the superimposition of digital objects on real-world images) makes it possible to visualize different future scenarios in order to choose the most desirable one, to begin a process of change. Agroforestry is an ideal case study: the introduction of trees profoundly modifies the appearance of plots and landscapes, so visualization tools would be very effective; and tree growth is slow and the consequences of the farmer’s choices are only revealed several decades later.
The aim of the thesis is to develop two AR applications, and to evaluate them with farmers, agricultural colleges and agroforestry advisors. The first will improve the design of agroforestry systems by allowing interactive comparison of configurations and visualization of the evolution of vegetation dynamics. This approach will be tested in agroforestry system design workshops. The second will realistically visualize in situ the growth of trees in agricultural plots, allowing the farmer to visit his plots and visualize their appearance 10, 20 or 40 years later. This research work aims to integrate growth models into AR technologies and the participatory approach to meet farmers’ needs.

Humanities and Social Sciences – Life and Environmental Sciences – Applied Mathematics

Pesticide use reduction, risks and value of digital information in agriculture

Key words : Pesticide use, decision-support tools, sensors, risk aversion, risk perception, information value

  • Contact : Marc Willinger – marc.willinger@umontpellier.fr
  • PhD Director(s): Douadia BOUGHERARA, CEE-M, INRAE – Marc WILLINGER, CEE-M, Université de Montpellier
  • Supervisor(s): Sophie Thoyer, CEE-M, INRAE – François Brun, ACTA – Emmanuelle Gourdain, Arvalis
  • Research units : CEE-M
  • Co-funding : Université de Montpellier
  • #DigitAg : Axe 1 : Impact des technologies de l’information et de la communication sur le monde rural, Axe 3 : Capteurs, acquisition et gestion de données, Axe 6 : Modélisation et simulation (systèmes de production agricole), Challenge 3 : La protection des cultures, Challenge 5 : Les services de conseil agricole

The many tools to support farmers’ decision to reduce pesticide use have often a low degree of adoption. Our hypothesis is that they are designed without taking into account farmers’ risk preferences and risk perceptions. We focus on the following question: how to improve our understanding of these preferences and to measure the value of new information (provided by sensors of digital agriculture) in order to design decision making tools that are more efficient and better fitted to farmers’ needs at the individual but also at the territory level? The question will be addressed in three steps using at least two contrasted case studies. (1) To define and measure farmers’ risk perceptions. In a multidisciplinary approach, we will characterize the risks associated with pesticides use reduction and consider a “multi-risk” approach. We will also elicit farmers’ perceptions and risk preferences for these risks using experimental economics methods on samples of farmers (field experiments); (2) To measure the value of digital information for farmers. We will develop a theoretical model of the determinants (socioeconomic, agronomic, ecological) of the value of information for farmers in the specific case of digital information decision making tools. We will then use experimental and/or statistical approaches to understand how new digital information can improve the value of the information provided by decision making tools; (3) Communicating risk using digital information decision making tools. We will determine how to optimally build and communicate information to users of digital information decision tools. Behavioral economics will be used. The results will help improve decision modelling and decision making tool design, and help orientate information collection, recording, tracking and sharing. Finally, from a more generic viewpoint, the thesis will allow an economic analysis of the issues raised by information platforms such as property rights issues (private/public) and its relationship with agricultural advice in general.

Engineering Science + Life & Environmental Sciences

Joint spectro-spatial analysis methods for the the in-situ health monitoring in orchards

Key words : Signal and image processing – spectroscopy\chemometrics

  • Contact : Florent Abdelghafour – florent.abdelghafour@irstea.fr
  • PhD Director(s) : Jean Michel Roger, ITAP, INRAE / Ryad Bendoula, ITAP, INRAE
  • Supervisor(s) : Florent Abdelghafour, ITAP, INRAE/ Florence Verpont, Unité mécanisation, CTIFL – UMT Ecotech-VITI-arbo / Nathalie Gorretta, ITAP, INRAE
  • Research units: ITAP
  • Co-funding : #DigitAg – Acta
  • #DigitAg : Axe 3 : Capteurs, acquisition et gestion de données, Challenge 3 : La protection des cultures, Challenge 6 : La gestion des territoires agricoles

Epidemiological surveillance is a crucial issue for food security, health safety and environmental protection. Perennial arboreal crops, such as apple and pear, are notably sensitive to phytopathologies, in particular “apple scab” and “fire blight”. Therefore, these issues are heightened in orchards where the substantial use of phytopharmaceutical inputs, the frequent exposure of operators and the vicinity with residents are sources of raising concerns or even conflicts.

In this context, Hyperspectral imaging (HIS), especially in proximal sensing, proves to be a relevant solution to describe and discriminate complex physiological phenomena involved in the diagnosis of phytopathologies. However, HIS is in practice difficult to implement for large-scale surveillance purposes. Indeed, numerous operating limits regarding instrumentation costs and more importantly regarding the measurements kinetic make HIS unpractical for industrial use. In order to sort these constraints, most of farm applications proceed by degrading the spectral resolution and then process simpler multi-spectral images (MSI). This strategy, although very functional in field, lacks of the spectral richness necessary to characterise phytopathologies.

The purpose of this thesis is to propose a strategy aiming at characterising in-situ the sanitary status of orchards at a local scale. This strategy is an alternative to MSI approaches consisting in estimating HIS from data acquired with more frugal sensors. Hence, it is proposed to investigate instrumentation and data analysis approaches aiming at combining rich spectral and spatial data in farming conditions. In practice, it consists in determining relevant combinations and adaptations of well-established chemometric and image processing methods or even developing hybrid methods dedicated to process altogether the spectral information, the geometrical, the statistical and the textural properties of images.

Engineering Science – Life & Environmental Sciences

Potentials of multi-year high-resolution Sentinel image time series for a better understanding of fallow dynamics in West Africa

Keywords: Tropical agricultural systems, fallows, land use, remote sensing, Sentinel, radar and optical imagery, multi-year time series

  • Contact: pierre-yves.vion [AT] agroparistech.fr
  • PhD director(s) : Agnès Bégué TETIS, Cirad
  • Supervisor(s): Raffaele Gaetano, TETIS Cirad/ Louise Leroux, AIDA Cirad
  • Research units: TETIS & AIDA
  • Co-funding :  #DigitAg – Cirad
  • #DigitAg: Axe 5 : Fouille de données, analyse de données, extraction de connaissances, Challenge 8 : Développement agricole au Sud, Axe 6 : Modélisation et simulation (systèmes de production agricole), Challenge 1 : Le challenge agroécologique,Challenge 4 : Des productions animales durables,Challenge 6 : La gestion des territoires agricoles

Extensive farming systems, still widespread in the tropics, are generally based on fallow practices, because of their ability to regenerate soil fertility, particularly through the maintenance of biomass reservoirs. Their importance has also been emphasized in adaptation to climate change, as they contribute to carbon sequestration and the reduction of greenhouse gases. As a result, the estimation of fallow areas is an important piece of information in assessing the performance of an agricultural system, both in terms of short-term productivity and the quantification of the “land stock” available for the establishment of strategies in response to climatic and/or anthropogenic factors. If the documentation of this practice in different regions of the world is important, a regular and exhaustive inventory of fallow land in West Africa does not exist. Given the stakes involved in this practice, this thesis aims to define a methodological framework combining expert knowledge and satellite imagery for the implementation of a fallow monitoring system at large scale. Indeed, fallow mapping is poorly taken into account in real-world remote sensing based land cover products. At best, this problem is naively approached omitting any consideration of the specificities related to these practices (extents, durations, strategies, but also their role in landscapes). In order to overcome these limitations, and with a growing availability of satellite images adapted to the monitoring of West African complex agricultural landscapes (such as those from ESA’s Sentinel missions), we will promote an interdisciplinary approach to (i) study the relationship between fallow land use and remote sensing indicators and (ii) to convey this information in the design of methods for analyzing multi-year series of images for their identification and characterization.

Life & Environmental Sciences – Engineering Science

Automated on real-time welfare and health assessment of gestating sows using heterogeneous data

Keywords: behaviour, welfare, precision breeding, technology, sows

  • Contact:  charlotte.gaillard [AT] inrae.fr
  • PhD director(s) : Jean-Yves Dourmand (INRAE PEGASE)
  • Supervisor(s): Charlotte Gaillard (INRAE UMR Pegase) et Christine Largouët (Inria, UMR IRISA-Inria)
  • Research units: INRAE Pegase/ Inria LACODAM (Rennes)
  • Co-funding : #DigitAg – INRAE
  • #DigitAg: Axes 5 & 2, Challenges 4 & 2

Evaluating sows’ behaviour allows the early detection of health or welfare problems and the quantification of their physical activity, which is major factor affecting their energy requirement. In practice, the continuous observation of each animal by the breeder is impossible and only intermittent observations are performed, often only once a day. New technologies have been developed like the accelerometers measuring the activity of each animal and its behaviour as well as software analysing animals’ vocalizations. Video analysis also offers the possibility to study social interactions between animals. Feeding and drinking behaviours may also be collected by feed and water dispensers. These behavioural data coupled with production data (feed intake, body weight and backfat thickness) should allow real-time estimation of the welfare of each animal, estimated through different criteria, and anticipate the occurrence of health problems in a non-invasive manner. The innovative issue is to describe the behavioural models using timed automata. The learning of timed automata is a rather new area of research. Their interest lies in time representation, the efficiency of exploration (using model-checking techniques) and the explainability of the models for predictive and monitoring tasks. The objectives of this thesis are therefore, in a first step, (i) to collect behavioural and production data in different situations (non-stressful vs. stressful), (ii) to learn and model the behaviours with timed automata describing gestating sows and the relationships between them in different situations (iii) to set up alerts and actions system to improve welfare, and (iv) to test in the experimental farm the capacity of this system to improve welfare and health.

Engineering Sciences- Life & Environmental Sciences

Signaling environmental-friendly practices through digitalisation; A farmer’s strategy to create value ?

Key words : Resources dependency theory, digitalization, signalisation, phytosanitary quality, short food supply chain, France, Italy

  • Contact : Anne Mione – anne.mione@umontpellier.fr
  • PhD Director(s) : Anne Mione, MRM, Université de Montpellier
  • Supervisor(s) : Magali Aubert, Moisa, Inrae – Fédérica de Maria, CREA
  • Research units: MRM
  • Co-funding : #DigitAg – Université de Montpellier
  • #DigitAg : Axe 1 : Impact des technologies de l’information et de la communication sur le monde rural, Axe 2 : Innovations en agriculture numérique, Challenge 7 : Intégration de l’agriculture dans les chaînes de valeur,  Challenge 3 : La protection des cultures

The project aims to analyze the role of digital tools in strengthening producer-consumer relations and promoting more environmentally friendly practices. It is in line with the European “farm to table” strategy of the Green Pact for Europe, which aims to offer healthy consumption to consumers by strengthening alternative practices. Based on economic and management science concepts, the approach aims to characterise the heterogeneity of behaviours in terms of adoption of digital tools and to associate them with producers’ productive and commercial strategies. The internship mission is to identify the endogeneous and exogeneous factors explaining the adoption of digital tools, based on resources theory. Concerning the PhD, we deepen the perspective and anchor the research on the resource dependence theory (Pfeffer et Salancik, 1978). Our hypothesis is that digitalisation enables the farmer to manage its dependency to environment. He better controls the production-commercial relation through direct sells and certification. From an empirical point of view, we mobilize longitudinal exhaustive data from all French and Italian farms (Census of Agriculture and structure survey). We know the level of digitalisation of producers, the certifications adopted and their marketing methods. In methodological terms, this project is based on the analysis of simultaneous triple equation models and matching type models. France and Italy are subject to the same requirements in terms of the reduction of phytosanitary products but have different commercial and production strategies. Comparing them makes it possible to assess the relationship between digitisation, certification and marketing in relation to the environment in which producers operate. These data will be supplemented by surveys of producers to study these relations more precisely with regard to existing digital tools and forms of production and marketing. The qualitative data will enable to appreciate how do farmers estimate their dependence to consumers, distributors, suppliers and complementers and to which extent do they consider that digitalization may modify their access to the environment resources.